current research topics in data science

Research Topics & Ideas: Data Science

50 Topic Ideas To Kickstart Your Research Project

If you’re just starting out exploring data science-related topics for your dissertation, thesis or research project, you’ve come to the right place. In this post, we’ll help kickstart your research by providing a hearty list of data science and analytics-related research ideas , including examples from recent studies.

PS – This is just the start…

We know it’s exciting to run through a list of research topics, but please keep in mind that this list is just a starting point . These topic ideas provided here are intentionally broad and generic , so keep in mind that you will need to develop them further. Nevertheless, they should inspire some ideas for your project.

To develop a suitable research topic, you’ll need to identify a clear and convincing research gap , and a viable plan to fill that gap. If this sounds foreign to you, check out our free research topic webinar that explores how to find and refine a high-quality research topic, from scratch. Alternatively, consider our 1-on-1 coaching service .

Data Science-Related Research Topics

• Developing machine learning models for real-time fraud detection in online transactions.
• The use of big data analytics in predicting and managing urban traffic flow.
• Investigating the effectiveness of data mining techniques in identifying early signs of mental health issues from social media usage.
• The application of predictive analytics in personalizing cancer treatment plans.
• Analyzing consumer behavior through big data to enhance retail marketing strategies.
• The role of data science in optimizing renewable energy generation from wind farms.
• Developing natural language processing algorithms for real-time news aggregation and summarization.
• The application of big data in monitoring and predicting epidemic outbreaks.
• Investigating the use of machine learning in automating credit scoring for microfinance.
• The role of data analytics in improving patient care in telemedicine.
• Developing AI-driven models for predictive maintenance in the manufacturing industry.
• The use of big data analytics in enhancing cybersecurity threat intelligence.
• Investigating the impact of sentiment analysis on brand reputation management.
• The application of data science in optimizing logistics and supply chain operations.
• Developing deep learning techniques for image recognition in medical diagnostics.
• The role of big data in analyzing climate change impacts on agricultural productivity.
• Investigating the use of data analytics in optimizing energy consumption in smart buildings.
• The application of machine learning in detecting plagiarism in academic works.
• Analyzing social media data for trends in political opinion and electoral predictions.
• The role of big data in enhancing sports performance analytics.
• Developing data-driven strategies for effective water resource management.
• The use of big data in improving customer experience in the banking sector.
• Investigating the application of data science in fraud detection in insurance claims.
• The role of predictive analytics in financial market risk assessment.
• Developing AI models for early detection of network vulnerabilities.

Data Science Research Ideas (Continued)

• The application of big data in public transportation systems for route optimization.
• Investigating the impact of big data analytics on e-commerce recommendation systems.
• The use of data mining techniques in understanding consumer preferences in the entertainment industry.
• Developing predictive models for real estate pricing and market trends.
• The role of big data in tracking and managing environmental pollution.
• Investigating the use of data analytics in improving airline operational efficiency.
• The application of machine learning in optimizing pharmaceutical drug discovery.
• Analyzing online customer reviews to inform product development in the tech industry.
• The role of data science in crime prediction and prevention strategies.
• Developing models for analyzing financial time series data for investment strategies.
• The use of big data in assessing the impact of educational policies on student performance.
• Investigating the effectiveness of data visualization techniques in business reporting.
• The application of data analytics in human resource management and talent acquisition.
• Developing algorithms for anomaly detection in network traffic data.
• The role of machine learning in enhancing personalized online learning experiences.
• Investigating the use of big data in urban planning and smart city development.
• The application of predictive analytics in weather forecasting and disaster management.
• Analyzing consumer data to drive innovations in the automotive industry.
• The role of data science in optimizing content delivery networks for streaming services.
• Developing machine learning models for automated text classification in legal documents.
• The use of big data in tracking global supply chain disruptions.
• Investigating the application of data analytics in personalized nutrition and fitness.
• The role of big data in enhancing the accuracy of geological surveying for natural resource exploration.
• Developing predictive models for customer churn in the telecommunications industry.
• The application of data science in optimizing advertisement placement and reach.

Recent Data Science-Related Studies

While the ideas we’ve presented above are a decent starting point for finding a research topic, they are fairly generic and non-specific. So, it helps to look at actual studies in the data science and analytics space to see how this all comes together in practice.

Below, we’ve included a selection of recent studies to help refine your thinking. These are actual studies,  so they can provide some useful insight as to what a research topic looks like in practice.

• Data Science in Healthcare: COVID-19 and Beyond (Hulsen, 2022)
• Auto-ML Web-application for Automated Machine Learning Algorithm Training and evaluation (Mukherjee & Rao, 2022)
• Survey on Statistics and ML in Data Science and Effect in Businesses (Reddy et al., 2022)
• Visualization in Data Science VDS @ KDD 2022 (Plant et al., 2022)
• An Essay on How Data Science Can Strengthen Business (Santos, 2023)
• A Deep study of Data science related problems, application and machine learning algorithms utilized in Data science (Ranjani et al., 2022)
• You Teach WHAT in Your Data Science Course?!? (Posner & Kerby-Helm, 2022)
• Statistical Analysis for the Traffic Police Activity: Nashville, Tennessee, USA (Tufail & Gul, 2022)
• Data Management and Visual Information Processing in Financial Organization using Machine Learning (Balamurugan et al., 2022)
• A Proposal of an Interactive Web Application Tool QuickViz: To Automate Exploratory Data Analysis (Pitroda, 2022)
• Applications of Data Science in Respective Engineering Domains (Rasool & Chaudhary, 2022)
• Jupyter Notebooks for Introducing Data Science to Novice Users (Fruchart et al., 2022)
• Towards a Systematic Review of Data Science Programs: Themes, Courses, and Ethics (Nellore & Zimmer, 2022)
• Application of data science and bioinformatics in healthcare technologies (Veeranki & Varshney, 2022)
• TAPS Responsibility Matrix: A tool for responsible data science by design (Urovi et al., 2023)
• Data Detectives: A Data Science Program for Middle Grade Learners (Thompson & Irgens, 2022)
• MACHINE LEARNING FOR NON-MAJORS: A WHITE BOX APPROACH (Mike & Hazzan, 2022)
• COMPONENTS OF DATA SCIENCE AND ITS APPLICATIONS (Paul et al., 2022)
• Analysis on the Application of Data Science in Business Analytics (Wang, 2022)

As you can see, these research topics are a lot more focused than the generic topic ideas we presented earlier. So, for you to develop a high-quality research topic, you’ll need to get specific and laser-focused on a specific context with specific variables of interest.  In the video below, we explore some other important things you’ll need to consider when crafting your research topic.

Get 1-On-1 Help

If you’re still unsure about how to find a quality research topic, check out our Research Topic Kickstarter service, which is the perfect starting point for developing a unique, well-justified research topic.

You Might Also Like:

Submit a Comment Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

• Print Friendly

37 Research Topics In Data Science To Stay On Top Of

• February 22, 2024

As a data scientist, staying on top of the latest research in your field is essential.

The data science landscape changes rapidly, and new techniques and tools are constantly being developed.

To keep up with the competition, you need to be aware of the latest trends and topics in data science research.

In this article, we will provide an overview of 37 hot research topics in data science.

We will discuss each topic in detail, including its significance and potential applications.

These topics could be an idea for a thesis or simply topics you can research independently.

Stay tuned – this is one blog post you don’t want to miss!

37 Research Topics in Data Science

1.) predictive modeling.

Predictive modeling is a significant portion of data science and a topic you must be aware of.

Simply put, it is the process of using historical data to build models that can predict future outcomes.

Predictive modeling has many applications, from marketing and sales to financial forecasting and risk management.

As businesses increasingly rely on data to make decisions, predictive modeling is becoming more and more important.

While it can be complex, predictive modeling is a powerful tool that gives businesses a competitive advantage.

2.) Big Data Analytics

These days, it seems like everyone is talking about big data.

And with good reason – organizations of all sizes are sitting on mountains of data, and they’re increasingly turning to data scientists to help them make sense of it all.

But what exactly is big data? And what does it mean for data science?

Simply put, big data is a term used to describe datasets that are too large and complex for traditional data processing techniques.

Big data typically refers to datasets of a few terabytes or more.

But size isn’t the only defining characteristic – big data is also characterized by its high Velocity (the speed at which data is generated), Variety (the different types of data), and Volume (the amount of the information).

Given the enormity of big data, it’s not surprising that organizations are struggling to make sense of it all.

That’s where data science comes in.

Data scientists use various methods to wrangle big data, including distributed computing and other decentralized technologies.

With the help of data science, organizations are beginning to unlock the hidden value in their big data.

By harnessing the power of big data analytics, they can improve their decision-making, better understand their customers, and develop new products and services.

3.) Auto Machine Learning

Auto machine learning is a research topic in data science concerned with developing algorithms that can automatically learn from data without intervention.

This area of research is vital because it allows data scientists to automate the process of writing code for every dataset.

This allows us to focus on other tasks, such as model selection and validation.

Auto machine learning algorithms can learn from data in a hands-off way for the data scientist – while still providing incredible insights.

This makes them a valuable tool for data scientists who either don’t have the skills to do their own analysis or are struggling.

4.) Text Mining

Text mining is a research topic in data science that deals with text data extraction.

This area of research is important because it allows us to get as much information as possible from the vast amount of text data available today.

Text mining techniques can extract information from text data, such as keywords, sentiments, and relationships.

This information can be used for various purposes, such as model building and predictive analytics.

5.) Natural Language Processing

Natural language processing is a data science research topic that analyzes human language data.

This area of research is important because it allows us to understand and make sense of the vast amount of text data available today.

Natural language processing techniques can build predictive and interactive models from any language data.

Natural Language processing is pretty broad, and recent advances like GPT-3 have pushed this topic to the forefront.

6.) Recommender Systems

Recommender systems are an exciting topic in data science because they allow us to make better products, services, and content recommendations.

Businesses can better understand their customers and their needs by using recommender systems.

This, in turn, allows them to develop better products and services that meet the needs of their customers.

Recommender systems are also used to recommend content to users.

This can be done on an individual level or at a group level.

Think about Netflix, for example, always knowing what you want to watch!

Recommender systems are a valuable tool for businesses and users alike.

7.) Deep Learning

Deep learning is a research topic in data science that deals with artificial neural networks.

These networks are composed of multiple layers, and each layer is formed from various nodes.

Deep learning networks can learn from data similarly to how humans learn, irrespective of the data distribution.

This makes them a valuable tool for data scientists looking to build models that can learn from data independently.

The deep learning network has become very popular in recent years because of its ability to achieve state-of-the-art results on various tasks.

There seems to be a new SOTA deep learning algorithm research paper on  https://arxiv.org/  every single day!

8.) Reinforcement Learning

Reinforcement learning is a research topic in data science that deals with algorithms that can learn on multiple levels from interactions with their environment.

This area of research is essential because it allows us to develop algorithms that can learn non-greedy approaches to decision-making, allowing businesses and companies to win in the long term compared to the short.

9.) Data Visualization

Data visualization is an excellent research topic in data science because it allows us to see our data in a way that is easy to understand.

Data visualization techniques can be used to create charts, graphs, and other visual representations of data.

This allows us to see the patterns and trends hidden in our data.

Data visualization is also used to communicate results to others.

This allows us to share our findings with others in a way that is easy to understand.

There are many ways to contribute to and learn about data visualization.

Some ways include attending conferences, reading papers, and contributing to open-source projects.

10.) Predictive Maintenance

Predictive maintenance is a hot topic in data science because it allows us to prevent failures before they happen.

This is done using data analytics to predict when a failure will occur.

This allows us to take corrective action before the failure actually happens.

While this sounds simple, avoiding false positives while keeping recall is challenging and an area wide open for advancement.

11.) Financial Analysis

Financial analysis is an older topic that has been around for a while but is still a great field where contributions can be felt.

Current researchers are focused on analyzing macroeconomic data to make better financial decisions.

This is done by analyzing the data to identify trends and patterns.

Financial analysts can use this information to make informed decisions about where to invest their money.

Financial analysis is also used to predict future economic trends.

This allows businesses and individuals to prepare for potential financial hardships and enable companies to be cash-heavy during good economic conditions.

Overall, financial analysis is a valuable tool for anyone looking to make better financial decisions.

12.) Image Recognition

Image recognition is one of the hottest topics in data science because it allows us to identify objects in images.

This is done using artificial intelligence algorithms that can learn from data and understand what objects you’re looking for.

This allows us to build models that can accurately recognize objects in images and video.

This is a valuable tool for businesses and individuals who want to be able to identify objects in images.

Think about security, identification, routing, traffic, etc.

Image Recognition has gained a ton of momentum recently – for a good reason.

13.) Fraud Detection

Fraud detection is a great topic in data science because it allows us to identify fraudulent activity before it happens.

This is done by analyzing data to look for patterns and trends that may be associated with the fraud.

Once our machine learning model recognizes some of these patterns in real time, it immediately detects fraud.

This allows us to take corrective action before the fraud actually happens.

Fraud detection is a valuable tool for anyone who wants to protect themselves from potential fraudulent activity.

14.) Web Scraping

Web scraping is a controversial topic in data science because it allows us to collect data from the web, which is usually data you do not own.

This is done by extracting data from websites using scraping tools that are usually custom-programmed.

This allows us to collect data that would otherwise be inaccessible.

For obvious reasons, web scraping is a unique tool – giving you data your competitors would have no chance of getting.

I think there is an excellent opportunity to create new and innovative ways to make scraping accessible for everyone, not just those who understand Selenium and Beautiful Soup.

15.) Social Media Analysis

Social media analysis is not new; many people have already created exciting and innovative algorithms to study this.

However, it is still a great data science research topic because it allows us to understand how people interact on social media.

This is done by analyzing data from social media platforms to look for insights, bots, and recent societal trends.

Once we understand these practices, we can use this information to improve our marketing efforts.

For example, if we know that a particular demographic prefers a specific type of content, we can create more content that appeals to them.

Social media analysis is also used to understand how people interact with brands on social media.

This allows businesses to understand better what their customers want and need.

Overall, social media analysis is valuable for anyone who wants to improve their marketing efforts or understand how customers interact with brands.

16.) GPU Computing

GPU computing is a fun new research topic in data science because it allows us to process data much faster than traditional CPUs .

Due to how GPUs are made, they’re incredibly proficient at intense matrix operations, outperforming traditional CPUs by very high margins.

While the computation is fast, the coding is still tricky.

There is an excellent research opportunity to bring these innovations to non-traditional modules, allowing data science to take advantage of GPU computing outside of deep learning.

17.) Quantum Computing

Quantum computing is a new research topic in data science and physics because it allows us to process data much faster than traditional computers.

It also opens the door to new types of data.

There are just some problems that can’t be solved utilizing outside of the classical computer.

For example, if you wanted to understand how a single atom moved around, a classical computer couldn’t handle this problem.

You’ll need to utilize a quantum computer to handle quantum mechanics problems.

This may be the “hottest” research topic on the planet right now, with some of the top researchers in computer science and physics worldwide working on it.

You could be too.

18.) Genomics

Genomics may be the only research topic that can compete with quantum computing regarding the “number of top researchers working on it.”

Genomics is a fantastic intersection of data science because it allows us to understand how genes work.

This is done by sequencing the DNA of different organisms to look for insights into our and other species.

Once we understand these patterns, we can use this information to improve our understanding of diseases and create new and innovative treatments for them.

Genomics is also used to study the evolution of different species.

Genomics is the future and a field begging for new and exciting research professionals to take it to the next step.

19.) Location-based services

Location-based services are an old and time-tested research topic in data science.

Since GPS and 4g cell phone reception became a thing, we’ve been trying to stay informed about how humans interact with their environment.

This is done by analyzing data from GPS tracking devices, cell phone towers, and Wi-Fi routers to look for insights into how humans interact.

Once we understand these practices, we can use this information to improve our geotargeting efforts, improve maps, find faster routes, and improve cohesion throughout a community.

Location-based services are used to understand the user, something every business could always use a little bit more of.

While a seemingly “stale” field, location-based services have seen a revival period with self-driving cars.

20.) Smart City Applications

Smart city applications are all the rage in data science research right now.

By harnessing the power of data, cities can become more efficient and sustainable.

But what exactly are smart city applications?

In short, they are systems that use data to improve city infrastructure and services.

This can include anything from traffic management and energy use to waste management and public safety.

Data is collected from various sources, including sensors, cameras, and social media.

It is then analyzed to identify tendencies and habits.

This information can make predictions about future needs and optimize city resources.

As more and more cities strive to become “smart,” the demand for data scientists with expertise in smart city applications is only growing.

21.) Internet Of Things (IoT)

The Internet of Things, or IoT, is exciting and new data science and sustainability research topic.

IoT is a network of physical objects embedded with sensors and connected to the internet.

These objects can include everything from alarm clocks to refrigerators; they’re all connected to the internet.

That means that they can share data with computers.

And that’s where data science comes in.

Data scientists are using IoT data to learn everything from how people use energy to how traffic flows through a city.

They’re also using IoT data to predict when an appliance will break down or when a road will be congested.

Really, the possibilities are endless.

With such a wide-open field, it’s easy to see why IoT is being researched by some of the top professionals in the world.

22.) Cybersecurity

Cybersecurity is a relatively new research topic in data science and in general, but it’s already garnering a lot of attention from businesses and organizations.

After all, with the increasing number of cyber attacks in recent years, it’s clear that we need to find better ways to protect our data.

While most of cybersecurity focuses on infrastructure, data scientists can leverage historical events to find potential exploits to protect their companies.

Sometimes, looking at a problem from a different angle helps, and that’s what data science brings to cybersecurity.

Also, data science can help to develop new security technologies and protocols.

As a result, cybersecurity is a crucial data science research area and one that will only become more important in the years to come.

23.) Blockchain

Blockchain is an incredible new research topic in data science for several reasons.

First, it is a distributed database technology that enables secure, transparent, and tamper-proof transactions.

Did someone say transmitting data?

This makes it an ideal platform for tracking data and transactions in various industries.

Second, blockchain is powered by cryptography, which not only makes it highly secure – but is a familiar foe for data scientists.

Finally, blockchain is still in its early stages of development, so there is much room for research and innovation.

As a result, blockchain is a great new research topic in data science that vows to revolutionize how we store, transmit and manage data.

24.) Sustainability

Sustainability is a relatively new research topic in data science, but it is gaining traction quickly.

To keep up with this demand, The Wharton School of the University of Pennsylvania has  started to offer an MBA in Sustainability .

This demand isn’t shocking, and some of the reasons include the following:

Sustainability is an important issue that is relevant to everyone.

Datasets on sustainability are constantly growing and changing, making it an exciting challenge for data scientists.

There hasn’t been a “set way” to approach sustainability from a data perspective, making it an excellent opportunity for interdisciplinary research.

As data science grows, sustainability will likely become an increasingly important research topic.

25.) Educational Data

Education has always been a great topic for research, and with the advent of big data, educational data has become an even richer source of information.

By studying educational data, researchers can gain insights into how students learn, what motivates them, and what barriers these students may face.

Besides, data science can be used to develop educational interventions tailored to individual students’ needs.

Imagine being the researcher that helps that high schooler pass mathematics; what an incredible feeling.

With the increasing availability of educational data, data science has enormous potential to improve the quality of education.

26.) Politics

As data science continues to evolve, so does the scope of its applications.

Originally used primarily for business intelligence and marketing, data science is now applied to various fields, including politics.

By analyzing large data sets, political scientists (data scientists with a cooler name) can gain valuable insights into voting patterns, campaign strategies, and more.

Further, data science can be used to forecast election results and understand the effects of political events on public opinion.

With the wealth of data available, there is no shortage of research opportunities in this field.

As data science evolves, so does our understanding of politics and its role in our world.

27.) Cloud Technologies

Cloud technologies are a great research topic.

It allows for the outsourcing and sharing of computer resources and applications all over the internet.

This lets organizations save money on hardware and maintenance costs while providing employees access to the latest and greatest software and applications.

I believe there is an argument that AWS could be the greatest and most technologically advanced business ever built (Yes, I know it’s only part of the company).

Besides, cloud technologies can help improve team members’ collaboration by allowing them to share files and work on projects together in real-time.

As more businesses adopt cloud technologies, data scientists must stay up-to-date on the latest trends in this area.

By researching cloud technologies, data scientists can help organizations to make the most of this new and exciting technology.

28.) Robotics

Robotics has recently become a household name, and it’s for a good reason.

First, robotics deals with controlling and planning physical systems, an inherently complex problem.

Second, robotics requires various sensors and actuators to interact with the world, making it an ideal application for machine learning techniques.

Finally, robotics is an interdisciplinary field that draws on various disciplines, such as computer science, mechanical engineering, and electrical engineering.

As a result, robotics is a rich source of research problems for data scientists.

29.) HealthCare

Healthcare is an industry that is ripe for data-driven innovation.

Hospitals, clinics, and health insurance companies generate a tremendous amount of data daily.

This data can be used to improve the quality of care and outcomes for patients.

This is perfect timing, as the healthcare industry is undergoing a significant shift towards value-based care, which means there is a greater need than ever for data-driven decision-making.

As a result, healthcare is an exciting new research topic for data scientists.

There are many different ways in which data can be used to improve healthcare, and there is a ton of room for newcomers to make discoveries.

30.) Remote Work

There’s no doubt that remote work is on the rise.

In today’s global economy, more and more businesses are allowing their employees to work from home or anywhere else they can get a stable internet connection.

But what does this mean for data science? Well, for one thing, it opens up a whole new field of research.

For example, how does remote work impact employee productivity?

What are the best ways to manage and collaborate on data science projects when team members are spread across the globe?

And what are the cybersecurity risks associated with working remotely?

These are just a few of the questions that data scientists will be able to answer with further research.

So if you’re looking for a new topic to sink your teeth into, remote work in data science is a great option.

31.) Data-Driven Journalism

Data-driven journalism is an exciting new field of research that combines the best of both worlds: the rigor of data science with the creativity of journalism.

By applying data analytics to large datasets, journalists can uncover stories that would otherwise be hidden.

And telling these stories compellingly can help people better understand the world around them.

Data-driven journalism is still in its infancy, but it has already had a major impact on how news is reported.

In the future, it will only become more important as data becomes increasingly fluid among journalists.

It is an exciting new topic and research field for data scientists to explore.

32.) Data Engineering

Data engineering is a staple in data science, focusing on efficiently managing data.

Data engineers are responsible for developing and maintaining the systems that collect, process, and store data.

In recent years, there has been an increasing demand for data engineers as the volume of data generated by businesses and organizations has grown exponentially.

Data engineers must be able to design and implement efficient data-processing pipelines and have the skills to optimize and troubleshoot existing systems.

If you are looking for a challenging research topic that would immediately impact you worldwide, then improving or innovating a new approach in data engineering would be a good start.

33.) Data Curation

Data curation has been a hot topic in the data science community for some time now.

Curating data involves organizing, managing, and preserving data so researchers can use it.

Data curation can help to ensure that data is accurate, reliable, and accessible.

It can also help to prevent research duplication and to facilitate the sharing of data between researchers.

Data curation is a vital part of data science. In recent years, there has been an increasing focus on data curation, as it has become clear that it is essential for ensuring data quality.

As a result, data curation is now a major research topic in data science.

There are numerous books and articles on the subject, and many universities offer courses on data curation.

Data curation is an integral part of data science and will only become more important in the future.

34.) Meta-Learning

Meta-learning is gaining a ton of steam in data science. It’s learning how to learn.

So, if you can learn how to learn, you can learn anything much faster.

Meta-learning is mainly used in deep learning, as applications outside of this are generally pretty hard.

In deep learning, many parameters need to be tuned for a good model, and there’s usually a lot of data.

You can save time and effort if you can automatically and quickly do this tuning.

In machine learning, meta-learning can improve models’ performance by sharing knowledge between different models.

For example, if you have a bunch of different models that all solve the same problem, then you can use meta-learning to share the knowledge between them to improve the cluster (groups) overall performance.

I don’t know how anyone looking for a research topic could stay away from this field; it’s what the  Terminator  warned us about!

35.) Data Warehousing

A data warehouse is a system used for data analysis and reporting.

It is a central data repository created by combining data from multiple sources.

Data warehouses are often used to store historical data, such as sales data, financial data, and customer data.

This data type can be used to create reports and perform statistical analysis.

Data warehouses also store data that the organization is not currently using.

This type of data can be used for future research projects.

Data warehousing is an incredible research topic in data science because it offers a variety of benefits.

Data warehouses help organizations to save time and money by reducing the need for manual data entry.

They also help to improve the accuracy of reports and provide a complete picture of the organization’s performance.

Data warehousing feels like one of the weakest parts of the Data Science Technology Stack; if you want a research topic that could have a monumental impact – data warehousing is an excellent place to look.

36.) Business Intelligence

Business intelligence aims to collect, process, and analyze data to help businesses make better decisions.

Business intelligence can improve marketing, sales, customer service, and operations.

It can also be used to identify new business opportunities and track competition.

BI is business and another tool in your company’s toolbox to continue dominating your area.

Data science is the perfect tool for business intelligence because it combines statistics, computer science, and machine learning.

Data scientists can use business intelligence to answer questions like, “What are our customers buying?” or “What are our competitors doing?” or “How can we increase sales?”

Business intelligence is a great way to improve your business’s bottom line and an excellent opportunity to dive deep into a well-respected research topic.

37.) Crowdsourcing

One of the newest areas of research in data science is crowdsourcing.

Crowdsourcing is a process of sourcing tasks or projects to a large group of people, typically via the internet.

This can be done for various purposes, such as gathering data, developing new algorithms, or even just for fun (think: online quizzes and surveys).

But what makes crowdsourcing so powerful is that it allows businesses and organizations to tap into a vast pool of talent and resources they wouldn’t otherwise have access to.

And with the rise of social media, it’s easier than ever to connect with potential crowdsource workers worldwide.

Imagine if you could effect that, finding innovative ways to improve how people work together.

That would have a huge effect.

Final Thoughts, Are These Research Topics In Data Science For You?

Thirty-seven different research topics in data science are a lot to take in, but we hope you found a research topic that interests you.

If not, don’t worry – there are plenty of other great topics to explore.

The important thing is to get started with your research and find ways to apply what you learn to real-world problems.

We wish you the best of luck as you begin your data science journey!

Other Data Science Articles

We love talking about data science; here are a couple of our favorite articles:

• Why Are You Interested In Data Science?
• Recent Posts

• How Much MB is in 1 GB [You Won’t Believe the Answer!] - April 18, 2024
• How much does a software engineering manager at Twitch make? [Discover the Full Compensation Breakdown Here] - April 17, 2024
• Does Amazon Use Employee Monitoring Software? [Shocking Insights Revealed!] - April 17, 2024

Trending now

Research Areas

Main navigation.

The world is being transformed by data and data-driven analysis is rapidly becoming an integral part of science and society. Stanford Data Science is a collaborative effort across many departments in all seven schools. We strive to unite existing data science research initiatives and create interdisciplinary collaborations, connecting the data science and related methodologists with disciplines that are being transformed by data science and computation.

Our work supports research in a variety of fields where incredible advances are being made through the facilitation of meaningful collaborations between domain researchers, with deep expertise in societal and fundamental research challenges, and methods researchers that are developing next-generation computational tools and techniques, including:

Data Science for Wildland Fire Research

In recent years, wildfire has gone from an infrequent and distant news item to a centerstage isssue spanning many consecutive weeks for urban and suburban communities. Frequent wildfires are changing everyday lives for California in numerous ways -- from public safety power shutoffs to hazardous air quality -- that seemed inconceivable as recently as 2015. Moreover, elevated wildfire risk in the western United States (and similar climates globally) is here to stay into the foreseeable future. There is a plethora of problems that need solutions in the wildland fire arena; many of them are well suited to a data-driven approach.

Seminar Series

Data Science for Physics

Astrophysicists and particle physicists at Stanford and at the SLAC National Accelerator Laboratory are deeply engaged in studying the Universe at both the largest and smallest scales, with state-of-the-art instrumentation at telescopes and accelerator facilities

Data Science for Economics

Many of the most pressing questions in empirical economics concern causal questions, such as the impact, both short and long run, of educational choices on labor market outcomes, and of economic policies on distributions of outcomes. This makes them conceptually quite different from the predictive type of questions that many of the recently developed methods in machine learning are primarily designed for.

Data Science for Education

Educational data spans K-12 school and district records, digital archives of instructional materials and gradebooks, as well as student responses on course surveys. Data science of actual classroom interaction is also of increasing interest and reality.

Data Science for Human Health

It is clear that data science will be a driving force in transitioning the world’s healthcare systems from reactive “sick-based” care to proactive, preventive care.

Data Science for Humanity

Our modern era is characterized by massive amounts of data documenting the behaviors of individuals, groups, organizations, cultures, and indeed entire societies. This wealth of data on modern humanity is accompanied by massive digitization of historical data, both textual and numeric, in the form of historic newspapers, literary and linguistic corpora, economic data, censuses, and other government data, gathered and preserved over centuries, and newly digitized, acquired, and provisioned by libraries, scholars, and commercial entities.

Data Science for Linguistics

The impact of data science on linguistics has been profound. All areas of the field depend on having a rich picture of the true range of variation, within dialects, across dialects, and among different languages. The subfield of corpus linguistics is arguably as old as the field itself and, with the advent of computers, gave rise to many core techniques in data science.

Data Science for Nature and Sustainability

Many key sustainability issues translate into decision and optimization problems and could greatly benefit from data-driven decision making tools. In fact, the impact of modern information technology has been highly uneven, mainly benefiting large firms in profitable sectors, with little or no benefit in terms of the environment. Our vision is that data-driven methods can — and should — play a key role in increasing the efficiency and effectiveness of the way we manage and allocate our natural resources.

Ethics and Data Science

With the emergence of new techniques of machine learning, and the possibility of using algorithms to perform tasks previously done by human beings, as well as to generate new knowledge, we again face a set of new ethical questions.

The Science of Data Science

The practice of data analysis has changed enormously. Data science needs to find new inferential paradigms that allow data exploration prior to the formulation of hypotheses.

DiscoverDataScience.org

Data Science Trends 2023

Created by aasif.faizal

Indeed, based on the exponential rate at which technology is currently advancing, each year is likely to bring even more change than the previous. 2023 is likely to be no different, with several pieces of significant data science news that any big data professional should take note of.

This article will survey some of the biggest pieces of data science news headed our way in the year to come.

Included in this Article:

• Key Themes in Data Science

Data Analytics

Artificial intelligence, data science jobs.

• Cloud Based Operations

Data Visualization Advancements

• Deepfake Video and Audio

Python Growth

Cybersecurity.

• Additional Resources

2023 Trends in Data Science: An Overview of Key Themes

If the 21st century has shown us anything, it’s that big data is only going to continue to get… well, bigger. In fact, even the name “big data” makes reference to the ever-expanding nature of information technology that allows for more and more valuable data to be captured and interpreted. Over the last thirty years especially, we’ve witnessed the way that these insights can transform how entire industries operate, from enhancing marketing research and development to identifying areas for improvement in a company’s production model.

But if big data has been trending for so long already, what developments is it likely to undergo in the year to come? Experts have identified a few of the following as key concerns for big data professionals in 2023:

• Advancing tools. Developments in A.I. and machine learning are certain to continue to make leaps in the coming year, which in turn will greatly impact the instruments data scientists have at their disposal to perform their research and analysis. This has already significantly impacted most branches of data science, all of which employ machine learning tools to function. In part, this is because of the next factor on the list…
• Higher volumes of data. As a direct result of the advances in A.I. and M.L. listed above, businesses are receiving a significantly larger amount of data of all sorts. Some of this expands upon previously existing datasets while other advances have brought in new forms of data altogether. In both cases, adaptations are required to allow businesses to make use of this data as well as finding safe and affordable ways to store it. As you will see, this will be critical in businesses across industries.
• Security threats. For better or worse, these are a perennial trend in the world of data science, with new forms of cyberattack emerging constantly. For those working in the field of cybersecurity , this means strategic planning and rigorous research to identify data breaches as well as preventing new ones. For everyone else working in big data, it means that vigilance is always imperative, as certain types of cyberattacks like phishing and other forms of data related fraud involve deceptive tactics that could target them.

Read on to discover our list of the specific data science trends of 2023 for insight into where we are and where we’re headed.

Analytics is one of the key fields of data science likely to undergo major transformations in the year ahead. Below are a few of the new data analytics trends to look out for.

Real-Time Analytics

One of the top rising data trends of 2023 is real-time analytics. Data capturing tools have improved in speed and scope, meaning we have access to an even greater wealth of real-time information that can illuminate our understanding of all sorts of processes. Companies are only just at the beginning of learning how these data sets can be used to guide important business decisions. If you are a data analyst or work in a related field, it will be of great use to follow any news and updates about real-time analytics that arise over the course of this year.

Mobile Analytics

Though it might seem like it couldn’t possibly continue to grow, the presence of mobile devices is ever-expanding in communities all over the globe. This means that there is significantly more mobile data to be captured. For certain industries, mobile analytics are the core of their strategy, providing the most revealing and useful information to guide marketing and advertising tactics. This includes registering user engagement, customer satisfaction, monitoring in-app traffic, and identifying security threats.

Recent years have made clear that artificial intelligence has made monumental leaps that are likely to change not only the way we do business but the way we live. Outside of data science, this has already risen to popular consciousness: A.I. programs that augment or generate images and texts have already begun to trend even among those who don’t know much about information technology.

Below are a few of the ways that artificial intelligence is likely to continue to trend in 2023.

Augmented Analytics

Within the field of big data, one of the main ways artificial intelligence is being employed is to help collect the ever-increasing amounts of data being captured and stored by new devices. This is in step with the global population’s increased dependence on technology to go about their everyday lives.

In order to keep up with the huge amounts of information that are continuously coming in, machine learning and AI tools are improving their processing functions to expedite this process, preparing and meaningfully analyzing a tremendous amount of new data. This is known as Augmented Analytics, and it can help businesses vastly.

Augmented analytics also fits into the category of Business Intelligence. The field is expected to grow at a breathtaking speed in the years to come.

Deep Learning

When people think of A.I., what they imagine is most comparable to deep learning, the branch of artificial intelligence devoted to training computers to behave like humans. These respond to neural network architectures and large sets of data that so far have been shown to be demonstrably effective. Indeed, in 2023 data scientists and casual observers alike are likely to take note of the increasing ability of computers to mirror human interactions.

For businesses, deep learning can be used to anticipate human behaviors, which in turn can impact areas including marketing and overall business strategies. As this technology grows more sophisticated, it will only continue to transform our approach to customer service and impact how businesses understand their customers’ needs.

For those who are considering breaking into the field of big data, here’s some exciting data science news: according to the Bureau of Labor Statistics, careers in big data are likely to be trending for years to come. This translates to more job openings and higher overall salaries, as data scientists are becoming ever-more valued by employers in all different sorts of industries.

Indeed, the numbers for data scientists are impressive by any metric: the Bureau of Labor Statistics reports an estimated job growth rate of an astonishing 36% by 2031, which is significantly higher than estimates for other professions. Over the years, different regions across the United States have become hubs for the tech industry, meaning an especially high concentration of data science job opportunities are available. Statistics from May 2021 reveal that the states with the highest employment levels for data scientists are California, New York, Texas, North Carolina, and Illinois.

In step with job growth projections, average data science salaries are exceptionally high relative to other professions. The Bureau of Labor Statistics reports a median annual salary of $100,910, with those in the highest-earning industry of scientific research and development earning a median annual salary of $102,750.

Businesses And Enterprises Moving To Cloud-Based Operations

In step with the increased amounts of data being captured by advanced tools, there is a growing need for enhanced storage solutions. Cloud computing is quickly being embraced as the solution to this problem, offering vastly improved storage opportunities that can keep up pace with the changing state of data capturing. In fact, most are beginning to consider it the very environment in which data-based business will be stored moving forward, though other storage options are technically available.

Of cloud-based computing service options available, the current game-changer is hybrid cloud, which makes use of machine learning and A.I. technology to offer a centralized database that is more cost-efficient than private cloud solutions (which can be all but completely out of reach in terms of cost for smaller businesses) and more secure than public cloud options. The hybrid cloud option is likely to become even more popular in 2023.

Cloud computing impacts the field of big tech in an enormous variety of ways, touching the fields of data science, customer interactions, Artificial Intelligence, transactional systems, DevOps, and more. If you are interested in or employed in any of those fields, staying abreast of developments in cloud-based technology will be crucial to managing your database efficiently and preserving it for the long term.

Several advancements in the field of data visualization are likely to be notable in 2023. Below are a few of the most important pieces of data science news to follow.

Data Visualization Videos

Those who are interested in pursuing careers in data visualization will be interested to learn about the trend toward using video for data visualization. Opting to use video instead of photos or text has been shown to significantly increase not only viewers’ engagement while watching but also increase their levels of fact retention in the days and even weeks after reviewing the data.

The central goal of data visualization is to help translate highly inaccessible information to be not comprehensible for business leaders who will make key decisions based on their findings. Because of this, sharing data sets through an entirely new medium that has been proven to be effective could be game changing for the field, expanding our capacity to learn at the same time that we expand our access to data.

Mobile-Optimized Visualization

Those who work in the arena of data visualization must be mindful not only of not only the medium in which they are working but the interface on which it will be seen. For those who provide data visualizations for remote clients, it’s important to remember that they may often review data visualizations on their mobile devices, meaning your visualization methods must be optimized to suit phones and tablets in addition to computers. Though it may sound simple, ease of access is a crucial factor in establishing long-term relationships with your clients, which means it’s imperative to confirm that your visualization products are reaching them quickly and straightforwardly. This is another data science trend to look out for in 2023.

Data as a Service (DaaS)

Having already begun to blossom, the field of DaaS is likely to continue to expand. For those not familiar with the term, Data as a Service is the industry in which organizations that have worked with data for decades share their expertise as well as their intellectual property with clients. This is because they possess uncommon insight into the workings of data that many data scientists seek to cultivate.

Because of the data industry’s ever-growing boom, Data as a Service is all but guaranteed to become an even more widespread industry, offering a great number of job opportunities to those who wish to specialize in the field. This is yet another piece of data science news to stay on the lookout for, as it’s likely to become a crucial part of data management.

Deepfake Video And Audio

Deepfake technology is another example of a tech-world development that has captured attention far outside of the field of big data. This is technology that can vividly create audio or video content by manipulating existing documents. Already, convincing video and audio clips have surfaced featuring public figures saying or doing things that they did not actually do, with highly destructive consequences. Indeed, the implications of this are vast, posing threats not only to the reputations of individuals but also causing destructive political misunderstandings.

Beyond these public examples of deepfake technology being weaponized, businesses in particular should be on the lookout for deepfake scams. Amazing though it may seem, the speech patterns and voices of individuals can be learned and mimicked by machines and used in automated cyberattacks. One famous example of this was in 2019, when a U.K.-based energy company was scammed out of close to a quarter million euros by a fraudulent phone call made using deepfake software to imitate the voice of a top-ranking executive.

Unfortunately, concern about deepfake technology will only continue to grow as machines continue to improve, becoming more convincing in what they are able to represent, quicker in their execution, and accurate in their responsiveness when interacting with real individuals. For those interested in the field of cybersecurity , this is an important new type of cyberattack to consider.

Python has long been one of the leading programming languages in data science, and in recent years has become the standard programming language for those in the field of data analysis. This is only likely to increase in the coming year, thanks to its relative ease of use (it is many beginners’ first programming language), its large number of data science libraries and machine learning libraries, and its availability for use in designing blockchain applications. Because of this, experts suggest that we are likely to see Python become the #1 programming language overall for data scientists , outpacing the other leading programming languages (javascript, java, and r).

If you are interested in pursuing a career in the data sciences, the message here is clear: if you have a choice of which programming language to learn first, Python is your strongest choice. Because of its rising popularity, it is possible that your degree program will also make teaching Python a priority for new students.

Below are a few of the key issues in cybersecurity that data scientists will face in the year ahead.

Adversarial Machine Learning Combating

Among the key targets of recent cyberattacks have been machine learning algorithms – the same system that has been radicalizing the world of big data in a whole variety of ways mentioned above. To prevent this, cybersecurity experts are working overtime to study adversarial machine learning to understand how these attacks work and what they can do to help prevent them in the future.

Indeed, there is reason to fear that these systems are under threat. A recent study by the Institute of Electrical and Electronic Engineers (IEEE) speaks in no uncertain terms, stating, “industry practitioners are not equipped with tactical and strategic tools to protect, detect and respond to attacks on their Machine Learning (M.L.) systems.” They argue that this is because data science research has so far been out of step with the rapidly growing capacities of M.L., which, being so instrumental to so many big data functions, is a major target for cyberattacks. Their study reveals that these systems themselves are not just attractive to cyberattackers but highly vulnerable, as they are insufficiently understood by data scientists.

For those interested in pursuing careers in cybersecurity , advances in adversarial machine learning is a piece of data science news to follow in 2023, as it is likely to make a lasting impact on many parts of the cybersecurity profession.

Consumer Data Protection

Security scandals have been a huge part of data science news over the years, in particular the Cambridge Analytica scandal , which exposed the illegal harvesting of private data accessed through individuals’ Facebook accounts to inform political campaigning. Indeed, data breaches of this nature are a huge source of anxiety and outrage among the public. This expands the job of cybersecurity experts whose efforts will not only help protect the companies they work for but will also contribute to the overall data safety of the public.

While cybersecurity experts will help figure out best practices for creating security programs as well as strategies to prevent increasingly sophisticated cyberattacks, there will also be significant efforts made in the realm of data science law to curb organizations’ and individuals’ ease of accessing data that is meant to be private.

Additional Resources for a Career in Data Science

If reading our Data Science Trends 2023 list is making you excited about the many developments in big data that are already transforming our world, you may be a perfect candidate to pursue a data science degree. As the article above may have made clear, there is a huge variety of career paths available in the world of data science, and choosing one early will help you gain the education and specialized skills you need to build a thriving career.

To learn more about the many career paths you can take in the world of big data, take a look at our comprehensive guide here .

One of the most popular specializations for data scientists is the field of data analytics, which itself can lead to a huge number of more specific focus areas. To learn more about the world of analytics and discover if it may be the right path for you, visit our guide to a career in analytics here .

After determining your area of focus, the next step is to find the data science program that is right for you. Data scientists can hold a huge variety of degrees and certifications, though the most common route to a high-earning data science career is through a master’s degree program. To learn more about educational opportunities and find the one that is right for you, visit our data science program guide here .

Finally, if you have great abilities with numbers but aren’t certain which industry is the right one to plot your career, we can help you survey the many options available to you. Our guide to exploring a career with numbers will give you the lay of the land so you can make a decision that will make the most of your interests and skills.

• Related Programs

Ten Research Challenge Areas in Data Science

Although data science builds on knowledge from computer science, mathematics, statistics, and other disciplines, data science is a unique field with many mysteries to unlock: challenging scientific questions and pressing questions of societal importance.

Is data science a discipline?

Data science is a field of study: one can get a degree in data science, get a job as a data scientist, and get funded to do data science research.  But is data science a discipline, or will it evolve to be one, distinct from other disciplines?  Here are a few meta-questions about data science as a discipline.

• What is/are the driving deep question(s) of data science?   Each scientific discipline (usually) has one or more “deep” questions that drive its research agenda: What is the origin of the universe (astrophysics)?  What is the origin of life (biology)?  What is computable (computer science)?  Does data science inherit its deep questions from all its constituency disciplines or does it have its own unique ones?
• What is the role of the domain in the field of data science?   People (including this author) (Wing, J.M., Janeia, V.P., Kloefkorn, T., & Erickson, L.C. (2018)) have argued that data science is unique in that it is not just about methods, but about the use of those methods in the context of a domain—the domain of the data being collected and analyzed; the domain for which a question to be answered comes from collecting and analyzing the data.  Is the inclusion of a domain inherent in defining the field of data science?  If so, is the way it is included unique to data science?
• What makes data science data science?   Is there a problem unique to data science that one can convincingly argue would not be addressed or asked by any of its constituent disciplines, e.g., computer science and statistics?

Ten research areas

While answering the above meta-questions is still under lively debate, including within the pages of this  journal, we can ask an easier question, one that also underlies any field of study: What are the research challenge areas that drive the study of data science?  Here is a list of ten.  They are not in any priority order, and some of them are related to each other.  They are phrased as challenge areas, not challenge questions.  They are not necessarily the “top ten” but they are a good ten to start the community discussing what a broad research agenda for data science might look like. 1

• Scientific understanding of learning, especially deep learning algorithms.    As much as we admire the astonishing successes of deep learning, we still lack a scientific understanding of why deep learning works so well.  We do not understand the mathematical properties of deep learning models.  We do not know how to explain why a deep learning model produces one result and not another.  We do not understand how robust or fragile they are to perturbations to input data distributions.  We do not understand how to verify that deep learning will perform the intended task well on new input data.  Deep learning is an example of where experimentation in a field is far ahead of any kind of theoretical understanding.
• Causal reasoning.   Machine learning is a powerful tool to find patterns and examine correlations, particularly in large data sets. While the adoption of machine learning has opened many fruitful areas of research in economics, social science, and medicine, these fields require methods that move beyond correlational analyses and can tackle causal questions. A rich and growing area of current study is revisiting causal inference in the presence of large amounts of data.  Economists are already revisiting causal reasoning by devising new methods at the intersection of economics and machine learning that make causal inference estimation more efficient and flexible (Athey, 2016), (Taddy, 2019).  Data scientists are just beginning to explore multiple causal inference, not just to overcome some of the strong assumptions of univariate causal inference, but because most real-world observations are due to multiple factors that interact with each other (Wang & Blei, 2018).
• Precious data.    Data can be precious for one of three reasons: the dataset is expensive to collect; the dataset contains a rare event (low signal-to-noise ratio );  or the dataset is artisanal—small and task-specific.   A good example of expensive data comes from large, one-of, expensive scientific instruments, e.g., the Large Synoptic Survey Telescope, the Large Hadron Collider, the IceCube Neutrino Detector at the South Pole.  A good example of rare event data is data from sensors on physical infrastructure, such as bridges and tunnels; sensors produce a lot of raw data, but the disastrous event they are used to predict is (thankfully) rare.   Rare data can also be expensive to collect.  A good example of artisanal data is the tens of millions of court judgments that China has released online to the public since 2014 (Liebman, Roberts, Stern, & Wang, 2017) or the 2+ million US government declassified documents collected by Columbia’s  History Lab  (Connelly, Madigan, Jervis, Spirling, & Hicks, 2019).   For each of these different kinds of precious data, we need new data science methods and algorithms, taking into consideration the domain and intended uses of the data.
• Multiple, heterogeneous data sources.   For some problems, we can collect lots of data from different data sources to improve our models.  For example, to predict the effectiveness of a specific cancer treatment for a human, we might build a model based on 2-D cell lines from mice, more expensive 3-D cell lines from mice, and the costly DNA sequence of the cancer cells extracted from the human. State-of-the-art data science methods cannot as yet handle combining multiple, heterogeneous sources of data to build a single, accurate model.  Since many of these data sources might be precious data, this challenge is related to the third challenge.  Focused research in combining multiple sources of data will provide extraordinary impact.
• Inferring from noisy and/or incomplete data.   The real world is messy and we often do not have complete information about every data point.  Yet, data scientists want to build models from such data to do prediction and inference.  A great example of a novel formulation of this problem is the planned use of differential privacy for Census 2020 data (Garfinkel, 2019), where noise is deliberately added to a query result, to maintain the privacy of individuals participating in the census. Handling “deliberate” noise is particularly important for researchers working with small geographic areas such as census blocks, since the added noise can make the data uninformative at those levels of aggregation. How then can social scientists, who for decades have been drawing inferences from census data, make inferences on this “noisy” data and how do they combine their past inferences with these new ones? Machine learning’s ability to better separate noise from signal can improve the efficiency and accuracy of those inferences.
• Trustworthy AI.   We have seen rapid deployment of systems using artificial intelligence (AI) and machine learning in critical domains such as autonomous vehicles, criminal justice, healthcare, hiring, housing, human resource management, law enforcement, and public safety, where decisions taken by AI agents directly impact human lives. Consequently, there is an increasing concern if these decisions can be trusted to be correct, reliable, robust, safe, secure, and fair, especially under adversarial attacks. One approach to building trust is through providing explanations of the outcomes of a machine learned model.  If we can interpret the outcome in a meaningful way, then the end user can better trust the model.  Another approach is through formal methods, where one strives to prove once and for all a model satisfies a certain property.  New trust properties yield new tradeoffs for machine learned models, e.g., privacy versus accuracy; robustness versus efficiency. There are actually multiple audiences for trustworthy models: the model developer, the model user, and the model customer.  Ultimately, for widespread adoption of the technology, it is the public who must trust these automated decision systems.
• Computing systems for data-intensive applications.    Traditional designs of computing systems have focused on computational speed and power: the more cycles, the faster the application can run.  Today, the primary focus of applications, especially in the sciences (e.g., astronomy, biology, climate science, materials science), is data.  Also, novel special-purpose processors, e.g., GPUs, FPGAs, TPUs, are now commonly found in large data centers. Even with all these data and all this fast and flexible computational power, it can still take weeks to build accurate predictive models; however, applications, whether from science or industry, want  real-time  predictions.  Also, data-hungry and compute-hungry algorithms, e.g., deep learning, are energy hogs (Strubell, Ganesh, & McCallum, 2019).   We should consider not only space and time, but also energy consumption, in our performance metrics.  In short, we need to rethink computer systems design from first principles, with data (not compute) the focus.  New computing systems designs need to consider: heterogeneous processing; efficient layout of massive amounts of data for fast access; the target domain, application, or even task; and energy efficiency.
• Automating front-end stages of the data life cycle.   While the excitement in data science is due largely to the successes of machine learning, and more specifically deep learning, before we get to use machine learning methods, we need to prepare the data for analysis.  The early stages in the data life cycle (Wing, 2019) are still labor intensive and tedious.  Data scientists, drawing on both computational and statistical methods, need to devise automated methods that address data cleaning and data wrangling, without losing other desired properties, e.g., accuracy, precision, and robustness, of the end model.One example of emerging work in this area is the Data Analysis Baseline Library (Mueller, 2019), which provides a framework to simplify and automate data cleaning, visualization, model building, and model interpretation.  The Snorkel project addresses the tedious task of data labeling (Ratner et al., 2018).
• Privacy.   Today, the more data we have, the better the model we can build.  One way to get more data is to share data, e.g., multiple parties pool their individual datasets to build collectively a better model than any one party can build.  However, in many cases, due to regulation or privacy concerns, we need to preserve the confidentiality of each party’s dataset.  An example of this scenario is in building a model to predict whether someone has a disease or not. If multiple hospitals could share their patient records, we could build a better predictive model; but due to Health Insurance Portability and Accountability Act (HIPAA) privacy regulations, hospitals cannot share these records. We are only now exploring practical and scalable ways, using cryptographic and statistical methods, for multiple parties to share data and/or share models to preserve the privacy of each party’s dataset.  Industry and government are exploring and exploiting methods and concepts, such as secure multi-party computation, homomorphic encryption, zero-knowledge proofs, and differential privacy, as part of a point solution to a point problem.
• Ethics.   Data science raises new ethical issues. They can be framed along three axes: (1) the ethics of data: how data are generated, recorded, and shared; (2) the ethics of algorithms: how artificial intelligence, machine learning, and robots interpret data; and (3) the ethics of practices: devising responsible innovation and professional codes to guide this emerging science (Floridi & Taddeo, 2016) and for defining Institutional Review Board (IRB) criteria and processes specific for data (Wing, Janeia, Kloefkorn, & Erickson 2018). Example ethical questions include how to detect and eliminate racial, gender, socio-economic, or other biases in machine learning models.

Closing remarks

As many universities and colleges are creating new data science schools, institutes, centers, etc. (Wing, Janeia, Kloefkorn, & Erickson 2018), it is worth reflecting on data science as a field.  Will data science as an area of research and education evolve into being its own discipline or be a field that cuts across all other disciplines?  One could argue that computer science, mathematics, and statistics share this commonality: they are each their own discipline, but they each can be applied to (almost) every other discipline. What will data science be in 10 or 50 years?

Acknowledgements

I would like to thank Cliff Stein, Gerad Torats-Espinosa, Max Topaz, and Richard Witten for their feedback on earlier renditions of this article.  Many thanks to all Columbia Data Science faculty who have helped me formulate and discuss these ten (and other) challenges during our Fall 2019 retreat.

Athey, S. (2016). “Susan Athey on how economists can use machine learning to improve policy,”  Retrieved from  https://siepr.stanford.edu/news/susan-athey-how-economists-can-use-machine-learning-improve-policy

Berger, J., He, X., Madigan, C., Murphy, S., Yu, B., & Wellner, J. (2019), Statistics at a Crossroad: Who is for the Challenge? NSF workshop report.  Retrieved from  https://hub.ki/groups/statscrossroad

Connelly, M., Madigan, D., Jervis, R., Spirling, A., & Hicks, R. (2019). The History Lab.  Retrieved from   http://history-lab.org/

Floridi , L. &  Taddeo , M. (2016). What is Data Ethics?  Philosophical Transactions of the Royal Society A , vol. 374, issue 2083, December 2016.

Garfinkel, S. (2019). Deploying Differential Privacy for the 2020 Census of Population and Housing. Privacy Enhancing Technologies Symposium, Stockholm, Sweden.  Retrieved from  http://simson.net/ref/2019/2019-07-16%20Deploying%20Differential%20Privacy%20for%20the%202020%20Census.pdf

Liebman, B.L., Roberts, M., Stern, R.E., & Wang, A. (2017).  Mass Digitization of Chinese Court Decisions: How to Use Text as Data in the Field of Chinese Law. UC  San Diego School of Global Policy and Strategy, 21 st  Century China Center Research Paper No. 2017-01; Columbia Public Law Research Paper No. 14-551. Retrieved from  https://scholarship.law.columbia.edu/faculty_scholarship/2039

Mueller, A. (2019). Data Analysis Baseline Library. Retrieved from  https://libraries.io/github/amueller/dabl

Ratner, A., Bach, S., Ehrenberg, H., Fries, J., Wu, S, & Ré, C. (2018).  Snorkel: Rapid Training Data Creation with Weak Supervision . Proceedings of the 44 th  International Conference on Very Large Data Bases.

Strubell E., Ganesh, A., & McCallum, A. (2019),”Energy and Policy Considerations for Deep Learning in NLP.  Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL).

Taddy, M. (2019).   Business Data Science: Combining Machine Learning and Economics to Optimize, Automate, and Accelerate Business Decisions , Mc-Graw Hill.

Wang, Y. & Blei, D.M. (2018). The Blessings of Multiple Causes, Retrieved from  https://arxiv.org/abs/1805.06826

Wing, J.M. (2019), The Data Life Cycle,  Harvard Data Science Review , vol. 1, no. 1. 

Wing, J.M., Janeia, V.P., Kloefkorn, T., & Erickson, L.C. (2018). Data Science Leadership Summit, Workshop Report, National Science Foundation.  Retrieved from  https://dl.acm.org/citation.cfm?id=3293458

J.M. Wing, “ Ten Research Challenge Areas in Data Science ,” Voices, Data Science Institute, Columbia University, January 2, 2020.  arXiv:2002.05658 .

Jeannette M. Wing is Avanessians Director of the Data Science Institute and professor of computer science at Columbia University.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Review Article
• Open access
• Published: 05 April 2022

Recent advances and applications of deep learning methods in materials science

• Kamal Choudhary   ORCID: orcid.org/0000-0001-9737-8074 1 , 2 , 3 ,
• Brian DeCost   ORCID: orcid.org/0000-0002-3459-5888 4 ,
• Chi Chen   ORCID: orcid.org/0000-0001-8008-7043 5 ,
• Anubhav Jain   ORCID: orcid.org/0000-0001-5893-9967 6 ,
• Francesca Tavazza   ORCID: orcid.org/0000-0002-5602-180X 1 ,
• Ryan Cohn   ORCID: orcid.org/0000-0002-7898-0059 7 ,
• Cheol Woo Park 8 ,
• Alok Choudhary 9 ,
• Ankit Agrawal 9 ,
• Simon J. L. Billinge   ORCID: orcid.org/0000-0002-9734-4998 10 ,
• Elizabeth Holm 7 ,
• Shyue Ping Ong   ORCID: orcid.org/0000-0001-5726-2587 5 &
• Chris Wolverton   ORCID: orcid.org/0000-0003-2248-474X 8  

npj Computational Materials volume  8 , Article number:  59 ( 2022 ) Cite this article

64k Accesses

221 Citations

38 Altmetric

Metrics details

• Atomistic models
• Computational methods

Deep learning (DL) is one of the fastest-growing topics in materials data science, with rapidly emerging applications spanning atomistic, image-based, spectral, and textual data modalities. DL allows analysis of unstructured data and automated identification of features. The recent development of large materials databases has fueled the application of DL methods in atomistic prediction in particular. In contrast, advances in image and spectral data have largely leveraged synthetic data enabled by high-quality forward models as well as by generative unsupervised DL methods. In this article, we present a high-level overview of deep learning methods followed by a detailed discussion of recent developments of deep learning in atomistic simulation, materials imaging, spectral analysis, and natural language processing. For each modality we discuss applications involving both theoretical and experimental data, typical modeling approaches with their strengths and limitations, and relevant publicly available software and datasets. We conclude the review with a discussion of recent cross-cutting work related to uncertainty quantification in this field and a brief perspective on limitations, challenges, and potential growth areas for DL methods in materials science.

Similar content being viewed by others

Machine-learned potentials for next-generation matter simulations

Pascal Friederich, Florian Häse, … Alán Aspuru-Guzik

AtomSets as a hierarchical transfer learning framework for small and large materials datasets

Chi Chen & Shyue Ping Ong

Automated classification of big X-ray diffraction data using deep learning models

Jerardo E. Salgado, Samuel Lerman, … Niaz Abdolrahim

Introduction

“Processing-structure-property-performance” is the key mantra in Materials Science and Engineering (MSE) 1 . The length and time scales of material structures and phenomena vary significantly among these four elements, adding further complexity 2 . For instance, structural information can range from detailed knowledge of atomic coordinates of elements to the microscale spatial distribution of phases (microstructure), to fragment connectivity (mesoscale), to images and spectra. Establishing linkages between the above components is a challenging task.

Both experimental and computational techniques are useful to identify such relationships. Due to rapid growth in automation in experimental equipment and immense expansion of computational resources, the size of public materials datasets has seen exponential growth. Several large experimental and computational datasets 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 have been developed through the Materials Genome Initiative (MGI) 11 and the increasing adoption of Findable, Accessible, Interoperable, Reusable (FAIR) 12 principles. Such an outburst of data requires automated analysis which can be facilitated by machine learning (ML) techniques 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 .

Deep learning (DL) 21 , 22 is a specialized branch of machine learning (ML). Originally inspired by biological models of computation and cognition in the human brain 23 , 24 , one of DL’s major strengths is its potential to extract higher-level features from the raw input data.

DL applications are rapidly replacing conventional systems in many aspects of our daily lives, for example, in image and speech recognition, web search, fraud detection, email/spam filtering, financial risk modeling, and so on. DL techniques have been proven to provide exciting new capabilities in numerous fields (such as playing Go 25 , self-driving cars 26 , navigation, chip design, particle physics, protein science, drug discovery, astrophysics, object recognition 27 , etc).

Recently DL methods have been outperforming other machine learning techniques in numerous scientific fields, such as chemistry, physics, biology, and materials science 20 , 28 , 29 , 30 , 31 , 32 . DL applications in MSE are still relatively new, and the field has not fully explored its potential, implications, and limitations. DL provides new approaches for investigating material phenomena and has pushed materials scientists to expand their traditional toolset.

DL methods have been shown to act as a complementary approach to physics-based methods for materials design. While large datasets are often viewed as a prerequisite for successful DL applications, techniques such as transfer learning, multi-fidelity modelling, and active learning can often make DL feasible for small datasets as well 33 , 34 , 35 , 36 .

Traditionally, materials have been designed experimentally using trial and error methods with a strong dose of chemical intuition. In addition to being a very costly and time-consuming approach, the number of material combinations is so huge that it is intractable to study experimentally, leading to the need for empirical formulation and computational methods. While computational approaches (such as density functional theory, molecular dynamics, Monte Carlo, phase-field, finite elements) are much faster and cheaper than experiments, they are still limited by length and time scale constraints, which in turn limits their respective domains of applicability. DL methods can offer substantial speedups compared to conventional scientific computing, and, for some applications, are reaching an accuracy level comparable to physics-based or computational models.

Moreover, entering a new domain of materials science and performing cutting-edge research requires years of education, training, and the development of specialized skills and intuition. Fortunately, we now live in an era of increasingly open data and computational resources. Mature, well-documented DL libraries make DL research much more easily accessible to newcomers than almost any other research field. Testing and benchmarking methodologies such as underfitting/overfitting/cross-validation 15 , 16 , 37 are common knowledge, and standards for measuring model performance are well established in the community.

Despite their many advantages, DL methods have disadvantages too, the most significant one being their black-box nature 38 which may hinder physical insights into the phenomena under examination. Evaluating and increasing the interpretability and explainability of DL models remains an active field of research. Generally a DL model has a few thousand to millions of parameters, making model interpretation and direct generation of scientific insight difficult.

Although there are several good recent reviews of ML applications in MSE 15 , 16 , 17 , 19 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , DL for materials has been advancing rapidly, warranting a dedicated review to cover the explosion of research in this field. This article discusses some of the basic principles in DL methods and highlights major trends among the recent advances in DL applications for materials science. As the tools and datasets for DL applications in materials keep evolving, we provide a github repository ( https://github.com/deepmaterials/dlmatreview ) that can be updated as new resources are made publicly available.

General machine learning concepts

It is beyond the scope of this article to give a detailed hands-on introduction to Deep Learning. There are many materials for this purpose, for example, the free online book “Neural Networks and Deep Learning” by Michael Nielsen ( http://neuralnetworksanddeeplearning.com ), Deep Learning by Goodfellow et al. 21 , and multiple online courses at Coursera, Udemy, and so on. Rather, this article aims to motivate materials scientist researchers in the types of problems that are amenable to DL, and to introduce some of the basic concepts, jargon, and materials-specific databases and software (at the time of writing) as a helpful on-ramp to help get started. With this in mind, we begin with a very basic introduction to Deep learning.

Artificial intelligence (AI) 13 is the development of machines and algorithms that mimic human intelligence, for example, by optimizing actions to achieve certain goals. Machine learning (ML) is a subset of AI, and provides the ability to learn without explicitly being programmed for a given dataset such as playing chess, social network recommendation etc. DL, in turn, is the subset of ML that takes inspiration from biological brains and uses multilayer neural networks to solve ML tasks. A schematic of AI-ML-DL context and some of the key application areas of DL in the materials science and engineering field are shown in Fig. 1 .

Deep learning is considered a part of machine learning, which is contained in an umbrella term artificial intelligence.

Some of the commonly used ML technologies are linear regression, decision trees, and random forest in which generalized models are trained to learn coefficients/weights/parameters for a given dataset (usually structured i.e., on a grid or a spreadsheet).

Applying traditional ML techniques to unstructured data (such as pixels or features from an image, sounds, text, and graphs) is challenging because users have to first extract generalized meaningful representations or features themselves (such as calculating pair-distribution for an atomic structure) and then train the ML models. Hence, the process becomes time-consuming, brittle, and not easily scalable. Here, deep learning (DL) techniques become more important.

DL methods are based on artificial neural networks and allied techniques. According to the “universal approximation theorem” 50 , 51 , neural networks can approximate any function to arbitrary accuracy. However, it is important to note that the theorem doesn’t guarantee that the functions can be learnt easily 52 .

Neural networks

A perceptron or a single artificial neuron 53 is the building block of artificial neural networks (ANNs) and performs forward propagation of information. For a set of inputs [ x 1 ,  x 2 , . . . ,  x m ] to the perceptron, we assign floating number weights (and biases to shift wights) [ w 1 ,  w 2 , . . . ,  w m ] and then we multiply them correspondingly together to get a sum of all of them. Some of the common software packages allowing NN trainings are: PyTorch 54 , Tensorflow 55 , and MXNet 56 . Please note that certain commercial equipment, instruments, or materials are identified in this paper in order to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by NIST, nor is it intended to imply that the materials or equipment identified are necessarily the best available for the purpose.

Activation function

Activation functions (such as sigmoid, hyperbolic tangent (tanh), rectified linear unit (ReLU), leaky ReLU, Swish) are the critical nonlinear components that enable neural networks to compose many small building blocks to learn complex nonlinear functions. For example, the sigmoid activation maps real numbers to the range (0, 1); this activation function is often used in the last layer of binary classifiers to model probabilities. The choice of activation function can affect training efficiency as well as final accuracy 57 .

Loss function, gradient descent, and normalization

The weight matrices of a neural network are initialized randomly or obtained from a pre-trained model. These weight matrices are multiplied with the input matrix (or output from a previous layer) and subjected to a nonlinear activation function to yield updated representations, which are often referred to as activations or feature maps. The loss function (also known as an objective function or empirical risk) is calculated by comparing the output of the neural network and the known target value data. Typically, network weights are iteratively updated via stochastic gradient descent algorithms to minimize the loss function until the desired accuracy is achieved. Most modern deep learning frameworks facilitate this by using reverse-mode automatic differentiation 58 to obtain the partial derivatives of the loss function with respect to each network parameter through recursive application of the chain rule. Colloquially, this is also known as back-propagation.

Common gradient descent algorithms include: Stochastic Gradient Descent (SGD), Adam, Adagrad etc. The learning rate is an important parameter in gradient descent. Except for SGD, all other methods use adaptive learning parameter tuning. Depending on the objective such as classification or regression, different loss functions such as Binary Cross Entropy (BCE), Negative Log likelihood (NLLL) or Mean Squared Error (MSE) are used.

The inputs of a neural network are generally scaled i.e., normalized to have zero mean and unit standard deviation. Scaling is also applied to the input of hidden layers (using batch or layer normalization) to improve the stability of ANNs.

Epoch and mini-batches

A single pass of the entire training data is called an epoch, and multiple epochs are performed until the weights converge. In DL, datasets are usually large and computing gradients for the entire dataset and network becomes challenging. Hence, the forward passes are done with small subsets of the training data called mini-batches.

Underfitting, overfitting, regularization, and early stopping

During an ML training, the dataset is split into training, validation, and test sets. The test set is never used during the training process. A model is said to be underfitting if the model performs poorly on the training set and lacks the capacity to fully learn the training data. A model is said to overfit if the model performs too well on the training data but does not perform well on the validation data. Overfitting is controlled with regularization techniques such as L2 regularization, dropout, and early stopping 37 .

Regularization discourages the model from simply memorizing the training data, resulting in a model that is more generalizable. Overfitting models are often characterized by neurons that have weights with large magnitudes. L2 regularization reduces the possibility of overfitting by adding an additional term to the loss function that penalizes the large weight values, keeping the values of the weights and biases small during training. Another popular regularization is dropout 59 in which we randomly set the activations for an NN layer to zero during training. Similar to bagging 60 , the use of dropout brings about the same effect of training a collection of randomly chosen models which prevents the co-adaptations among the neurons, consequently reducing the likelihood of the model from overfitting. In early stopping, further epochs for training are stopped before the model overfits i.e., accuracy on the validation set flattens or decreases.

Convolutional neural networks

Convolutional neural networks (CNN) 61 can be viewed as a regularized version of multilayer perceptrons with a strong inductive bias for learning translation-invariant image representations. There are four main components in CNNs: (a) learnable convolution filterbanks, (b) nonlinear activations, (c) spatial coarsening (via pooling or strided convolution), (d) a prediction module, often consisting of fully connected layers that operate on a global instance representation.

In CNNs we use convolution functions with multiple kernels or filters with trainable and shared weights or parameters, instead of general matrix multiplication. These filters/kernels are matrices with a relatively small number of rows and columns that convolve over the input to automatically extract high-level local features in the form of feature maps. The filters slide/convolve (element-wise multiply) across the input with a fixed number of strides to produce the feature map and the information thus learnt is passed to the hidden/fully connected layers. Depending on the input data, these filters can be one, two, or three-dimensional.

Similar to the fully connected NNs, nonlinearities such as ReLU are then applied that allows us to deal with nonlinear and complicated data. The pooling operation preserves spatial invariance, downsamples and reduces the dimension of each feature map obtained after convolution. These downsampling/pooling operations can be of different types such as maximum-pooling, minimum-pooling, average pooling, and sum pooling. After one or more convolutional and pooling layers, the outputs are usually reduced to a one-dimensional global representation. CNNs are especially popular for image data.

Graph neural networks

Graphs and their variants.

Classical CNNs as described above are based on a regular grid Euclidean data (such as 2D grid in images). However, real-life data structures, such as social networks, segments of images, word vectors, recommender systems, and atomic/molecular structures, are usually non-Euclidean. In such cases, graph-based non-Euclidean data structures become especially important.

Mathematically, a graph G is defined as a set of nodes/vertices V , a set of edges/links, E and node features, X : G  = ( V ,  E ,  X ) 62 , 63 , 64 and can be used to represent non-Euclidean data. An edge is formed between a pair of two nodes and contains the relation information between the nodes. Each node and edge can have attributes/features associated with it. An adjacency matrix A is a square matrix indicating connections between the nodes or not in the form of 1 (connected) and 0 (unconnected). A graph can be of various types such as: undirected/directed, weighted/unweighted, homogeneous/heterogeneous, static/dynamic.

An undirected graph captures symmetric relations between nodes, while a directed one captures asymmetric relations such that A i j  ≠  A j i . In a weighted graph, each edge is associated with a scalar weight rather than just 1s and 0s. In a homogeneous graph, all the nodes represent instances of the same type, and all the edges capture relations of the same type while in a heterogeneous graph, the nodes and edges can be of different types. Heterogeneous graphs provide an easy interface for managing nodes and edges of different types as well as their associated features. When input features or graph topology vary with time, they are called dynamic graphs otherwise they are considered static. If a node is connected to another node more than once it is termed a multi-graph.

Types of GNNs

At present, GNNs are probably the most popular AI method for predicting various materials properties based on structural information 33 , 65 , 66 , 67 , 68 , 69 . Graph neural networks (GNNs) are DL methods that operate on graph domain and can capture the dependence of graphs via message passing between the nodes and edges of graphs. There are two key steps in GNN training: (a) we first aggregate information from neighbors and (b) update the nodes and/or edges. Importantly, aggregation is permutation invariant. Similar to the fully connected NNs, the input node features, X (with embedding matrix) are multiplied with the adjacency matrix and the weight matrices and then multiplied with the nonlinear activation function to provide outputs for the next layer. This method is called the propagation rule.

Based on the propagation rule and aggregation methodology, there could be different variants of GNNs such as Graph convolutional network (GCN) 70 , Graph attention network (GAT) 71 , Relational-GCN 72 , graph recurrent network (GRN) 73 , Graph isomerism network (GIN) 74 , and Line graph neural network (LGNN) 75 . Graph convolutional neural networks are the most popular GNNs.

Sequence-to-sequence models

Traditionally, learning from sequential inputs such as text involves generating a fixed-length input from the data. For example, the “bag-of-words” approach simply counts the number of instances of each word in a document and produces a fixed-length vector that is the size of the overall vocabulary.

In contrast, sequence-to-sequence models can take into account sequential/contextual information about each word and produce outputs of arbitrary length. For example, in named entity recognition (NER), an input sequence of words (e.g., a chemical abstract) is mapped to an output sequence of “entities” or categories where every word in the sequence is assigned a category.

An early form of sequence-to-sequence model is the recurrent neural network, or RNN. Unlike the fully connected NN architecture, where there is no connection between hidden nodes in the same layer, but only between nodes in adjacent layers, RNN has feedback connections. Each hidden layer can be unfolded and processed similarly to traditional NNs sharing the same weight matrices. There are multiple types of RNNs, of which the most common ones are: gated recurrent unit recurrent neural network (GRURNN), long short-term memory (LSTM) network, and clockwork RNN (CW-RNN) 76 .

However, all such RNNs suffer from some drawbacks, including: (i) difficulty of parallelization and therefore difficulty in training on large datasets and (ii) difficulty in preserving long-range contextual information due to the “vanishing gradient” problem. Nevertheless, as we will later describe, LSTMs have been successfully applied to various NER problems in the materials domain.

More recently, sequence-to-sequence models based on a “transformer” architecture, such as Google’s Bidirectional Encoder Representations from Transformers (BERT) model 77 , have helped address some of the issues of traditional RNNs. Rather than passing a state vector that is iterated word-by-word, such models use an attention mechanism to allow access to all previous words simultaneously without explicit time steps. This mechanism facilitates parallelization and also better preserves long-term context.

Generative models

While the above DL frameworks are based on supervised machine learning (i.e., we know the target or ground truth data such as in classification and regression) and discriminative (i.e., learn differentiating features between various datasets), many AI tasks are based on unsupervised (such as clustering) and are generative (i.e., aim to learn underlying distributions) 78 .

Generative models are used to (a) generate data samples similar to the training set with variations i.e., augmentation and for synthetic data, (b) learn good generalized latent features, (c) guide mixed reality applications such as virtual try-on. There are various types of generative models, of which the most common are: (a) variational encoders (VAE), which explicitly define and learn likelihood of data, (b) Generative adversarial networks (GAN), which learn to directly generate samples from model’s distribution, without defining any density function.

A VAE model has two components: namely encoder and decoder. A VAE’s encoder takes input from a target distribution and compresses it into a low-dimensional latent space. Then the decoder takes that latent space representation and reproduces the original image. Once the network is trained, we can generate latent space representations of various images, and interpolate between these before forwarding them through the decoder which produces new images. A VAE is similar to a principal component analysis (PCA) but instead of linear data assumption in PCA, VAEs work in nonlinear domain. A GAN model also has two components: namely generator, and discriminator. GAN’s generator generates fake/synthetic data that could fool the discriminator. Its discriminator tries to distinguish fake data from real ones. This process is also termed as “min-max two-player game.” We note that VAE models learn the hidden state distributions during the training process, while GAN’s hidden state distributions are predefined. Rather GAN generators serve to generate images that could fool the discriminator. These techniques are widely used for images and spectra and have also been recently applied to atomic structures.

Deep reinforcement learning

Reinforcement learning (RL) deals with tasks in which a computational agent learns to make decisions by trial and error. Deep RL uses DL into the RL framework, allowing agents to make decisions from unstructured input data 79 . In traditional RL, Markov decision process (MDP) is used in which an agent at every timestep takes action to receive a scalar reward and transitions to the next state according to system dynamics to learn policy in order to maximize returns. However, in deep RL, the states are high-dimensional (such as continuous images or spectra) which act as an input to DL methods. DRL architectures can be either model-based or model-free.

Scientific machine learning

The nascent field of scientific machine learning (SciML) 80 is creating new opportunities across all paradigms of machine learning, and deep learning in particular. SciML is focused on creating ML systems that incorporate scientific knowledge and physical principles, either directly in the specific form of the model or indirectly through the optimization algorithms used for training. This offers potential improvements in sample and training complexity, robustness (particularly under extrapolation), and model interpretability. One prominent theme can be found in ref. 57 . Such implementations usually involve applying multiple physics-based constraints while training a DL model 81 , 82 , 83 . One of the key challenges of universal function approximation is that a NN can quickly learn spurious features that have nothing to do with the features that a researcher could be actually interested in, within the data. In this sense, physics-based regularization can assist. Physics-based deep learning can also aid in inverse design problems, a challenging but important task 84 , 85 . On the flip side, deep Learning using Graph Neural Nets and symbolic regression (stochastically building symbolic expressions) has even been used to “discover” symbolic equations from data that capture known (and unknown) physics behind the data 86 , i.e., to deep learn a physics model rather than to use a physics model to constrain DL.

Overview of applications

Some aspects of successful DL application that require materials-science-specific considerations are:

acquiring large, balanced, and diverse datasets (often on the order of 10,000 data points or more),

determing an appropriate DL approach and suitable vector or graph representation of the input samples, and

selecting appropriate performance metrics relevant to scientific goals.

In the following sections we discuss some of the key areas of materials science in which DL has been applied with available links to repositories and datasets that help in the reproducibility and extensibility of the work. In this review we categorize materials science applications at a high level by the type of input data considered: 11 atomistic, 12 stoichiometric, 13 spectral, 14 image, and 15 text. We summarize prevailing machine learning tasks and their impact on materials research and development within each broad materials data modality.

Applications in atomistic representations

In this section, we provide a few examples of solving materials science problems with DL methods trained on atomistic data. The atomic structure of material usually consists of atomic coordinates and atomic composition information of material. An arbitrary number of atoms and types of elements in a system poses a challenge to apply traditional ML algorithms for atomistic predictions. DL-based methods are an obvious strategy to tackle this problem. There have been several previous attempts to represent crystals and molecules using fixed-size descriptors such as Coulomb matrix 87 , 88 , 89 , classical force field inspired descriptors (CFID) 90 , 91 , 92 , pair-distribution function (PRDF), Voronoi tessellation 93 , 94 , 95 . Recently graph neural network methods have been shown to surpass previous hand-crafted feature set 28 .

DL for atomistic materials applications include: (a) force-field development, (b) direct property predictions, (c) materials screening. In addition to the above points, we also elucidate upon some of the recent generative adversarial network and complimentary methods to atomistic aproaches.

Databases and software libraries

In Table 1 we provide some of the commonly used datasets used for atomistic DL models for molecules, solids, and proteins. We note that the computational methods used for different datasets are different and many of them are continuously evolving. Generally it takes years to generate such databases using conventional methods such as density functional theory; in contrast, DL methods can be used to make predictions with much reduced computational cost and reasonable accuracy.

Table 1 we provide DL software packages used for atomistic materials design. The type of models includes general property (GP) predictors and interatomic force fields (FF). The models have been demonstrated in molecules (Mol), solid-state materials (Sol), or proteins (Prot). For some force fields, high-performance large-scale implementations (LSI) that leverage paralleling computing exist. Some of these methods mainly used interatomic distances to build graphs while others use distances as well as bond-angle information. Recently, including bond angle within GNN has been shown to drastically improve the performance with comparable computational timings.

Force-field development

The first application includes the development of DL-based force fields (FF) 96 , 97 /interatomic potentials. Some of the major advantages of such applications are that they are very fast (on the order of hundreds to thousands times 64 ) for making predictions and solving the tenuous development of FFs, but the disadvantage is they still require a large dataset using computationally expensive methods to train.

Models such as Behler-Parrinello neural network (BPNN) and its variants 98 , 99 are used for developing interatomic potentials that can be used beyond just 0 K temperature and time-dependent behavior using molecular dynamics simulations such as for nanoparticles 100 . Such FF models have been developed for molecular systems, such as water, methane, and other organic molecules 99 , 101 as well as solids such as silicon 98 , sodium 102 , graphite 103 , and titania ( T i O 2 ) 104 .

While the above works are mainly based on NNs, there has also been the development of graph neural network force-field (GNNFF) framework 105 , 106 that bypasses both computational bottlenecks. GNNFF can predict atomic forces directly using automatically extracted structural features that are not only translationally invariant, but rotationally-covariant to the coordinate space of the atomic positions, i.e., the features and hence the predicted force vectors rotate the same way as the rotation of coordinates. In addition to the development of pure NN-based FFs, there have also been recent developments of combining traditional FFs such as bond-order potentials with NNs and ReaxFF with message passing neural network (MPNN) that can help mitigate the NNs issue for extrapolation 82 , 107 .

Direct property prediction from atomistic configurations

DL methods can be used to establish a structure-property relationship between atomic structure and their properties with high accuracy 28 , 108 . Models such as SchNet, crystal graph convolutional neural network (CGCNN), improved crystal graph convolutional neural network (iCGCNN), directional message passing neural network (DimeNet), atomistic line graph neural network (ALIGNN) and materials graph neural network (MEGNet) shown in Table 1 have been used to predict up to 50 properties of crystalline and molecular materials. These property datasets are usually obtained from ab-initio calculations. A schematic of such models shown in Fig. 2 . While SchNet, CGCNN, MEGNet are primarily based on atomic distances, iCGCNN, DimeNet, and ALIGNN models capture many-body interactions using GCNN.

a CGCNN model in which crystals are converted to graphs with nodes representing atoms in the unit cell and edges representing atom connections. Nodes and edges are characterized by vectors corresponding to the atoms and bonds in the crystal, respectively [Reprinted with permission from ref. 67 Copyright 2019 American Physical Society], b ALIGNN 65 model in which the convolution layer alternates between message passing on the bond graph and its bond-angle line graph. c MEGNet in which the initial graph is represented by the set of atomic attributes, bond attributes and global state attributes [Reprinted with permission from ref. 33 Copyright 2019 American Chemical Society] model, d iCGCNN model in which multiple edges connect a node to neighboring nodes to show the number of Voronoi neighbors [Reprinted with permission from ref. 122 Copyright 2019 American Physical Society].

Some of these properties include formation energies, electronic bandgaps, solar-cell efficiency, topological spin-orbit spillage, dielectric constants, piezoelectric constants, 2D exfoliation energies, electric field gradients, elastic modulus, Seebeck coefficients, power factors, carrier effective masses, highest occupied molecular orbital, lowest unoccupied molecular orbital, energy gap, zero-point vibrational energy, dipole moment, isotropic polarizability, electronic spatial extent, internal energy.

For instance, the current state-of-the-art mean absolute error for formation energy for solids at 0 K is 0.022 eV/atom as obtained by the ALIGNN model 65 . DL is also heavily being used for predicting catalytic behavior of materials such as the Open Catalyst Project 109 which is driven by the DL methods materials design. There is an ongoing effort to continuously improve the models. Usually energy-based models such as formation and total energies are more accurate than electronic property-based models such as bandgaps and power factors.

In addition to molecules and solids, property predictions models have also been used for bio-materials such as proteins, which can be viewed as large molecules. There have been several efforts for predicting protein-based properties, such as binding affinity 66 and docking predictions 110 .

There have also been several applications for identifying reasonable chemical space using DL methods such as autoencoders 111 and reinforcement learning 112 , 113 , 114 for inverse materials design. Inverse materials design with techniques such as GAN deals with finding chemical compounds with suitable properties and act as complementary to forward prediction models. While such concepts have been widely applied to molecular systems, 115 , recently these methods have been applied to solids as well 116 , 117 , 118 , 119 , 120 .

Fast materials screening

DFT-based high-throughput methods are usually limited to a few thousands of compounds and take a long time for calculations, DL-based methods can aid this process and allow much faster predictions. DL-based property prediction models mentioned above can be used for pre-screening chemical compounds. Hence, DL-based tools can be viewed as a pre-screening tool for traditional methods such as DFT. For example, Xie et al. used CGCNN model to screen stable perovskite materials 67 as well hierarchical visualization of materials space 121 . Park et al. 122 used iCGCNN to screen T h C r 2 S i 2 -type materials. Lugier et al. used DL methods to predict thermoelectric properties 123 . Rosen et al. 124 used graph neural network models to predict the bandgaps of metal-organic frameworks. DL for molecular materials has been used to predict technologically important properties such as aqueous solubility 125 and toxicity 126 .

It should be noted that the full atomistic representations and the associated DL models are only possible if the crystal structure and atom positions are available. In practice, the precise atom positions are only available from DFT structural relaxations or experiments, and are one of the goals for materials discovery instead of the starting point. Hence, alternative methods have been proposed to bypass the necessity for atom positions in building DL models. For example, Jain and Bligaard 127 proposed the atomic position-independent descriptors and used a CNN model to learn the energies of crystals. Such descriptors include information based only on the symmetry (e.g., space group and Wyckoff position). In principle, the method can be applied universally in all crystals. Nevertheless, the model errors tend to be much higher than graph-based models. Similar coarse-grained representation using Wyckoff representation was also used by Goodall et al. 128 . Alternatively, Zuo et al. 129 started from the hypothetical structures without precise atom positions, and used a Bayesian optimization method coupled with a MEGNet energy model as an energy evaluator to perform direct structural relaxation. Applying the Bayesian optimization with symmetry relaxation (BOWSR) algorithm successfully discovered ReWB (Pca2 1 ) and MoWC 2 (P6 3 /mmc) hard materials, which were then experimentally synthesized.

Applications in chemical formula and segment representations

One of the earliest applications for DL included SMILES for molecules, elemental fractions and chemical descriptors for solids, and sequence of protein names as descriptors. Such descriptors lack explicit inclusion of atomic structure information but are still useful for various pre-screening applications for both theoretical and experimental data.

SMILES and fragment representation

The simplified molecular-input line-entry system (SMILES) is a method to represent elemental and bonding for molecular structures using short American Standard Code for Information Interchange (ASCII) strings. SMILES can express structural differences including the chirality of compounds, making it more useful than a simply chemical formula. A SMILES string is a simple grid-like (1-D grid) structure that can represent molecular sequences such as DNA, macromolecules/polymers, protein sequences also 130 , 131 . In addition to the chemical constituents as in the chemical formula, bondings (such as double and triple bondings) are represented by special symbols (such as ’=’ and ’#’). The presence of a branch point indicated using a left-hand bracket “(” while the right-hand bracket “)” indicates that all the atoms in that branch have been taken into account. SMILES strings are represented as a distributed representation termed a SMILES feature matrix (as a sparse matrix), and then we can apply DL to the matrix similar to image data. The length of the SMILES matrix is generally kept fixed (such as 400) during training and in addition to the SMILES multiple elemental attributes and bonding attributes (such as chirality, aromaticity) can be used. Key DL tasks for molecules include (a) novel molecule design, (b) molecule screening.

Novel molecules with target properties can designed using VAE, GAN and RNN based methods 132 , 133 , 134 . These DL-generated molecules might not be physically valid, but the goal is to train the model to learn the patterns in SMILES strings such that the output resembles valid molecules. Then chemical intuitions can be further used to screen the molecules. DL for SMILES can also be used for molecularscreening such as to predict molecular toxicity. Some of the common SMILES datasets are: ZINC 135 , Tox21 136 , and PubChem 137 .

Due to the limitations to enforce the generation of valid molecular structures from SMILES, fragment-based models are developed such as DeepFrag and DeepFrag-K 138 , 139 . In fragment-based models, a ligand/receptor complex is removed and then a DL model is trained to predict the most suitable fragment substituent. A set of useful tools for SMILES and fragment representations are provided in Table 2 .

Chemical formula representation

There are several ways of using the chemical formula-based representations for building ML/DL models, beginning with a simple vector of raw elemental fractions 140 , 141 or of weight percentages of alloying compositions 142 , 143 , 144 , 145 , as well as more sophisticated hand-crafted descriptors or physical attributes to add known chemistry knowledge (e.g., electronegativity, valency, etc. of constituent elements) to the feature representations 146 , 147 , 148 , 149 , 150 , 151 . Statistical and mathematical operations such as average, max, min, median, mode, and exponentiation can be carried out on elemental properties of the constituent elements to get a set of descriptors for a given compound. The number of such composition-based features can range from a few dozens to a few hundreds. One of the commonly used representations that have been shown to work for a variety of different use-cases is the materials agnostic platform for informatics and exploration (MagPie) 150 . All these composition-based representations can be used with both traditional ML methods such as Random Forest as well as DL.

It is relevant to note that ElemNet 141 , which is a 17-layer neural network composed of fully connected layers and uses only raw elemental fractions as input, was found to significantly outperform traditional ML methods such as Random Forest, even when they were allowed to use more sophisticated physical attributes based on MagPie as input. Although no periodic table information was provided to the model, it was found to self-learn some interesting chemistry, like groups (element similarity) and charge balance (element interaction). It was also able to predict phase diagrams on unseen materials systems, underscoring the power of DL for representation learning directly from raw inputs without explicit feature extraction. Further increasing the depth of the network was found to adversely affect the model accuracy due to the vanishing gradient problem. To address this issue, Jha et al. 152 developed IRNet, which uses individual residual learning to allow a smoother flow of gradients and enable deeper learning for cases where big data is available. IRNet models were tested on a variety of big and small materials datasets, such as OQMD, AFLOW, Materials Project, JARVIS, using different vector-based materials representations (element fractions, MagPie, structural) and were found to not only successfully alleviate the vanishing gradient problem and enable deeper learning, but also lead to significantly better model accuracy as compared to plain deep neural networks and traditional ML techniques for a given input materials representation in the presence of big data 153 . Further, graph-based methods such as Roost 154 have also been developed which can outperform many similar techniques.

Such methods have been used for diverse DFT datasets mentioned above in Table 1 as well as experimental datasets such as SuperCon 155 , 156 for quick pre-screening applications. In terms of applications, they have been applied for predicting properties such as formation energy 141 , bandgap, and magnetization 152 , superconducting temperatures 156 , bulk, and shear modulus 153 . They have also been used for transfer learning across datasets for enhanced predictive accuracy on small data 34 , even for different source and target properties 157 , which is especially useful to build predictive models for target properties for which big source datasets may not be readily available.

There have been libraries of such descriptors developed such as MatMiner 151 and DScribe 158 . Some examples of such models are given in Table 2 . Such representations are especially useful for experimental datasets such as those for superconducting materials where the atomic structure is not tabulated. However, these representations cannot distinguish different polymorphs of a system with different point groups and space groups. It has been recently shown that although composition-based representations can help build ML/DL models to predict some properties like formation energy with remarkable accuracy, it does not necessarily translate to accurate predictions of other properties such as stability, when compared to DFT’s own accuracy 159 .

Spectral models

When electromagnetic radiation hits materials, the interaction between the radiation and matter measured as a function of the wavelength or frequency of the radiation produces a spectroscopic signal. By studying spectroscopy, researchers can gain insights into the materials’ composition, structural, and dynamic properties. Spectroscopic techniques are foundational in materials characterization. For instance, X-ray diffraction (XRD) has been used to characterize the crystal structure of materials for more than a century. Spectroscopic analysis can involve fitting quantitative physical models (for example, Rietveld refinement) or more empirical approaches such as fitting linear combinations of reference spectra, such as with x-ray absorption near-edge spectroscopy (XANES). Both approaches require a high degree of researcher expertise through careful design of experiments; specification, revision, and iterative fitting of physical models; or the availability of template spectra of known materials. In recent years, with the advances in high-throughput experiments and computational data, spectroscopic data has multiplied, giving opportunities for researchers to learn from the data and potentially displace the conventional methods in analyzing such data. This section covers emerging DL applications in various modes of spectroscopic data analysis, aiming to offer practice examples and insights. Some of the applications are shown in Fig. 3 .

a Predicting structure information from the X-ray diffraction 374 , Reprinted according to the terms of the CC-BY license 374 . Copyright 2020. b Predicting catalysis properties from computational electronic density of states data. Reprinted according to the terms of the CC-BY license 202 . Copyright 2021.

Currently, large-scale and element-diverse spectral data mainly exist in computational databases. For example, in ref. 160 , the authors calculated the infrared spectra, piezoelectric tensor, Born effective charge tensor, and dielectric response as a part of the JARVIS-DFT DFPT database. The Materials Project has established the largest computational X-ray absorption database (XASDb), covering the K-edge X-ray near-edge fine structure (XANES) 161 , 162 and the L-edge XANES 163 of a large number of material structures. The database currently hosts more than 400,000 K-edge XANES site-wise spectra and 90,000 L-edge XANES site-wise spectra of many compounds in the Materials Project. There are considerably fewer experimental XAS spectra, being on the order of hundreds, as seen in the EELSDb and the XASLib. Collecting large experimental spectra databases that cover a wide range of elements is a challenging task. Collective efforts focused on curating data extracted from different sources, as found in the RRUFF Raman, XRD and chemistry database 164 , the open Raman database 165 , and the SOP spectra library 166 . However, data consistency is not guaranteed. It is also now possible for contributors to share experimental data in a Materials Project curated database, MPContribs 167 . This database is supported by the US Department of Energy (DOE) providing some expectation of persistence. Entries can be kept private or published and are linked to the main materials project computational databases. There is an ongoing effort to capture data from DOE-funded synchrotron light sources ( https://lightsources.materialsproject.org/ ) into MPContribs in the future.

Recent advances in sources, detectors, and experimental instrumentation have made high-throughput measurements of experimental spectra possible, giving rise to new possibilities for spectral data generation and modeling. Such examples include the HTEM database 10 that contains 50,000 optical absorption spectra and the UV-Vis database of 180,000 samples from the Joint Center for Artificial Photosynthesis. Some of the common spectra databases for spectra data are shown in Table 3 . There are beginning to appear cloud-based software as a service platforms for high-throughput data analysis, for example, pair-distribution function (PDF) in the cloud ( https://pdfitc.org ) 168 which are backed by structured databases, where data can be kept private or made public. This transition to the cloud from data analysis software installed and run locally on a user’s computer will facilitate the sharing and reuse of data by the community.

Applications

Due to the widespread deployment of XRD across many materials technologies, XRD spectra became one of the first test grounds for DL models. Phase identification from XRD can be mapped into a classification task (assuming all phases are known) or an unsupervised clustering task. Unlike the traditional analysis of XRD data, where the spectra are treated as convolved, discrete peak positions and intensities, DL methods treat the data as a continuous pattern similar to an image. Unfortunately, a significant number of experimental XRD datasets in one place are not readily available at the moment. Nevertheless, extensive, high-quality crystal structure data makes creating simulated XRD trivial.

Park et al. 169 calculated 150,000 XRD patterns from the Inorganic Crystal Structure Database (ICSD) structural database 170 and then used CNN models to predict structural information from the simulated XRD patterns. The accuracies of the CNN models reached 81.14%, 83.83%, and 94.99% for space-group, extinction-group, and crystal-system classifications, respectively.

Liu et al. 95 obtained similar accuracies by using a CNN for classifying atomic pair-distribution function (PDF) data into space groups. The PDF is obtained by Fourier transforming XRD into real space and is particularly useful for studying the local and nanoscale structure of materials. In the case of the PDF, models were trained, validated, and tested on simulated data from the ICSD. However, the trained model showed excellent performance when given experimental data, something that can be a challenge in XRD data because of the different resolutions and line-shapes of the diffraction data depending on specifics of the sample and experimental conditions. The PDF seems to be more robust against these aspects.

Similarly, Zaloga et al. 171 also used the ICSD database for XRD pattern generation and CNN models to classify crystals. The models achieved 90.02% and 79.82% accuracy for crystal systems and space groups, respectively.

It should be noted that the ICSD database contains many duplicates, and such duplicates should be filtered out to avoid information leakage. There is also a large difference in the number of structures represented in each space group (the label) in the database resulting in data normalization challenges.

Lee et al. 172 developed a CNN model for phase identification from samples consisting of a mixture of several phases in a limited chemical space relevant for battery materials. The training data are mixed patterns consisting of 1,785,405 synthetic XRD patterns from the Sr-Li-Al-O phase space. The resulting CNN can not only identify the phases but also predict the compound fraction in the mixture. A similar CNN was utilized by Wang et al. 173 for fast identification of metal-organic frameworks (MOFs), where experimental spectral noise was extracted and then synthesized into the theoretical XRD for training data augmentation.

An alternative idea was proposed by Dong et al. 174 . Instead of recognizing only phases from the CNN, a proposed “parameter quantification network” (PQ-Net) was able to extract physico-chemical information. The PQ-Net yields accurate predictions for scale factors, crystallite size, and lattice parameters for simulated and experimental XRD spectra. The work by Aguiar et al. 175 took a step further and proposed a modular neural network architecture that enables the combination of diffraction patterns and chemistry data and provided a ranked list of predictions. The ranked list predictions provide user flexibility and overcome some aspects of overconfidence in model predictions. In practical applications, AI-driven XRD identification can be beneficial for high-throughput materials discovery, as shown by Maffettone et al. 176 . In their work, an ensemble of 50 CNN models was trained on synthetic data reproducing experimental variations (missing peaks, broadening, peaking shifting, noises). The model ensemble is capable of predicting the probability of each category label. A similar data augmentation idea was adopted by Oviedo et al. 177 , where experimental XRD data for 115 thin-film metal-halides were measured, and CNN models trained on the augmented XRD data achieved accuracies of 93% and 89% for classifying dimensionality and space group, respectively.

Although not a DL method, an unsupervised machine learning approach, non-negative matrix factorization (NMF), is showing great promise for yielding chemically relevant XRD spectra from time- or spatially-dependent sets of diffraction patterns. NMF is closely related to principle component analysis in that it takes a set of patterns as a matrix and then compresses the data by reducing the dimensionality by finding the most important components. In NMF a constraint is applied that all the components and their weights must be strictly positive. This often corresponds to a real physical situation (for example, spectra tend to be positive, as are the weights of chemical constituents). As a result, it appears that the mathematical decomposition often results in interpretable, physically meaningful, components and weights, as shown by Liu et al. for PDF data 178 . An extension of this showed that in a spatially resolved study, NMF could be used to extract chemically resolved differential PDFs (similar to the information in EXAFS) from non-chemically resolved PDF measurements 179 . NMF is very quick and easy to apply and can be applied to just about any set of spectra. It is likely to become widely used and is being implemented in the PDFitc.org website to make it more accessible to potential users.

Other than XRD, the XAS, Raman, and infrared spectra, also contain rich structure-dependent spectroscopic information about the material. Unlike XRD, where relatively simple theories and equations exist to relate structures to the spectral patterns, the relationships between general spectra and structures are somewhat elusive. This difficulty has created a higher demand for machine learning models to learn structural information from other spectra.

For instance, the case of X-ray absorption spectroscopy (XAS), including the X-ray absorption near-edge spectroscopy (XANES) and extended X-ray absorption fine structure (EXAFS), is usually used to analyze the structural information on an atomic level. However, the high signal-to-noise XANES region has no equation for data fitting. DL modeling of XAS data is fascinating and offers unprecedented insights. Timoshenko et al. used neural networks to predict the coordination numbers of Pt 180 and Cu 181 in nanoclusters from the XANES. Aside from the high accuracies, the neural network also offers high prediction speed and new opportunities for quantitative XANES analysis. Timoshenko et al. 182 further carried out a novel analysis of EXAFS using DL. Although EXAFS analysis has an explicit equation to fit, the study is limited to the first few coordination shells and on relatively ordered materials. Timoshenko et al. 182 first transformed the EXAFS data into 2D maps with a wavelet transform and then supplied the 2D data to a neural network model. The model can instantly predict relatively long-range radial distribution functions, offering in situ local structure analysis of materials. The advent of high-throughput XAS databases has recently unveiled more possibilities for machine learning models to be deployed using XAS data. For example, Zheng et al. 161 used an ensemble learning method to match and fast search new spectra in the XASDb. Later, the same authors showed that random forest models outperform DL models such as MLPs or CNNs in directly predicting atomic environment labels from the XANES spectra 183 . Similar approaches were also adopted by Torrisi et al. 184 In practical applications, Andrejevic et al. 185 used the XASDb data together with the topological materials database. They constructed CNN models to classify the topology of materials from the XANES and symmetry group inputs. The model correctly predicted 81% topological and 80% trivial cases and achieved 90% accuracy in material classes containing certain elements.

Raman, infrared, and other vibrational spectroscopies provide structural fingerprints and are usually used to discriminate and estimate the concentration of components in a mixture. For example, Madden et al. 186 have used neural network models to predict the concentration of illicit materials in a mixture using the Raman spectra. Interestingly, several groups have independently found that DL models outperform chemometrics analysis in vibrational spectroscopies 187 , 188 . For learning vibrational spectra, the number of training spectra is usually less than or on the order of the number of features (intensity points), and the models can easily overfit. Hence, dimensional reduction strategies are commonly used to compress the information dimension using, for example, principal component analysis (PCA) 189 , 190 . DL approaches do not have such concerns and offer elegant and unified solutions. For example, Liu et al. 191 applied CNN models to the Raman spectra in the RRUFF spectral database and show that CNN models outperform classical machine learning models such as SVM in classification tasks. More DL applications in vibrational spectral analysis can be found in a recent review by Yang et al. 192 .

Although most current DL work focuses on the inverse problem, i.e., predicting structural information from the spectra, some innovative approaches also solve the forward problems by predicting the spectra from the structure. In this case, the spectroscopy data can be viewed simply as a high-dimensional material property of the structure. This is most common in molecular science, where predicting the infrared spectra 193 , molecular excitation spectra 194 , is of particular interest. In the early 2000s, Selzer et al. 193 and Kostka et al. 195 attempted predicting the infrared spectra directly from the molecular structural descriptors using neural networks. Non-DL models can also perform such tasks to a reasonable accuracy 196 . For DL models, Chen et al. 197 used a Euclidean neural network (E(3)NN) to predict the phonon density of state (DOS) spectra 198 from atom positions and element types. The E(3)NN model captures symmetries of the crystal structures, with no need to perform data augmentation to achieve target invariances. Hence the E(3)NN model is extremely data-efficient and can give reliable DOS spectra prediction and heat capacity using relatively sparse data of 1200 calculation results on 65 elements. A similar idea was also used to predict the XAS spectra. Carbone et al. 199 used a message passing neural network (MPNN) to predict the O and N K-edge XANES spectra from the molecular structures in the QM9 database 7 . The training XANES data were generated using the FEFF package 200 . The trained MPNN model reproduced all prominent peaks in the predicted XANES, and 90% of the predicted peaks are within 1 eV of the FEFF calculations. Similarly, Rankine et al. 201 started from the two-body radial distribution function (RDC) and used a deep neural network model to predict the Fe K-edge XANES spectra for arbitrary local environments.

In addition to learn the structure-spectra or spectra-structure relationships, a few works have also explored the possibility of relating spectra to other material properties in a non-trivial way. The DOSnet proposed by Fung et al. 202 (Fig. 3 b) uses the electronic DOS spectra calculated from DFT as inputs to a CNN model to predict the adsorption energies of H, C, N, O, S and their hydrogenated counterparts, CH, CH 2 , CH 3 , NH, OH, and SH, on bimetallic alloy surfaces. This approach extends the previous d-band theory 203 , where only the d-band center, a scalar, was used to correlate with the adsorption energy on transition metals. Similarly, Kaundinya et al. 204 used Atomistic Line Graph Neural Network (ALIGNN) to predict DOS for 56,000 materials in the JARVIS-DFT database using a direct discretized spectrum (D-ALIGNN), and a compressed low-dimensional representation using an autoencoder (AE-ALIGNN). Stein et al. 205 tried to learn the mapping between the image and the UV-vis spectrum of the material using the conditional variational encoder (cVAE) with neural network models as the backbone. Such models can generate the UV-vis spectrum directly from a simple material image, offering much faster material characterizations. Predicting gas adsorption isotherms for direct air capture (DAC) are also an important application of spectra-based DL models. There have been several important works 206 , 207 for CO 2 capture with high-performance metal-organic frameworks (MOFs) which are important for mitigating climate change issues.

Image-based models

Computer vision is often credited as precipitating the current wave of mainstream DL applications a decade ago 208 . Naturally, materials researchers have developed a broad portfolio of applications of computer vision for accelerating and improving image-based material characterization techniques. High-level microscopy vision tasks can be organized as follows: image classification (and material property regression), auto-tuning experimental imaging hyperparameters, pixelwise learning (e.g., semantic segmentation), super-resolution imaging, object/entity recognition, localization, and tracking, microstructure representation learning.

Often these tasks generalize across many different imaging modalities, spanning optical microscopy (OM), scanning electron microscopy (SEM) techniques, scanning probe microscopy (SPM, as in scanning tunneling microscopy (STM) or atomic force microscopy (AFM), and transmission electron microscopy (TEM) variants, including scanning transmission electron microscopy (STEM).

The images obtained with these techniques range from capturing local atomic to mesoscale structures (microstructure), the distribution and type of defects, and their dynamics which are critically linked to the functionality and performance of the materials. Over the past few decades, atomic-scale imaging has become widespread and near-routine due to aberration-corrected STEM 209 . The collection of large image datasets is increasingly presenting an analysis bottleneck in the materials characterization pipeline, and the immediate need for automated image analysis becomes important. Non-DL image analysis methods have driven tremendous progress in quantitative microscopy, but often image processing pipelines are brittle and require too much manual identification of image features to be broadly applicable. Thus, DL is currently the most promising solution for high-performance, high-throughput automated analysis of image datasets. For a good overview of applications in microstructure characterization specifically, see 210 .

Image datasets for materials can come from either experiments or simulations. Software libraries mentioned above can be used to generate images such as STM/STEM. Images can also be obtained from the literature. A few common examples for image datasets are shown below in Table 4 . Recently, there has been a rapid development in the field of image learning tasks for materials leading to several useful packages. We list some of them in Table 4 .

Applications in image classification and regression

DL for images can be used to automatically extract information from images or transform images into a more useful state. The benefits of automated image analysis include higher throughput, better consistency of measurements compared to manual analysis, and even the ability to measure signals in images that humans cannot detect. The benefits of altering images include image super-resolution, denoising, inferring 3D structure from 2D images, and more. Examples of the applications of each task are summarized below.

Image classification and regression

Classification and regression are the processes of predicting one or more values associated with an image. In the context of DL the only difference between the two methods is that the outputs of classification are discrete while the outputs of regression models are continuous. The same network architecture may be used for both classification and regression by choosing the appropriate activation function (i.e., linear for regression or Softmax for classification) for the output of the network. Due to its simplicity image classification is one of the most established DL techniques available in the materials science literature. Nonetheless, this technique remains an area of active research.

Modarres et al. applied DL with transfer learning to automatically classify SEM images of different material systems 211 . They demonstrated how a single approach can be used to identify a wide variety of features and material systems such as particles, fibers, Microelectromechanical systems (MEMS) devices, and more. The model achieved 90% accuracy on a test set. Misclassifications resulted from images containing objects from multiple classes, which is an inherent limitation of single-class classification. More advanced techniques such as those described in subsequent sections can be applied to avoid these limitations. Additionally, they developed a system to deploy the trained model at scale to process thousands of images in parallel. This approach is essential for large-scale, high-throughput experiments or industrial applications of classification. ImageNet-based deep transfer learning has also been successfully applied for crack detection in macroscale materials images 212 , 213 , as well as for property prediction on small, noisy, and heterogeneous industrial datasets 214 , 215 .

DL has also been applied to characterize the symmetries of simulated measurements of samples. In ref. 216 , Ziletti et al. obtained a large database of perfect crystal structures, introduced defects into the perfect lattices, and simulated diffraction patterns for each structure. DL models were trained to identify the space group of each diffraction patterns. The model achieved high classification performance, even on crystals with significant numbers of defects, surpassing the performance of conventional algorithms for detecting symmetries from diffraction patterns.

DL has also been applied to classify symmetries in simulated STM measurements of 2D material systems 217 . DFT was used to generate simulated STM images for a variety of material systems. A convolutional neural network was trained to identify which of the five 2D Bravais lattices each material belonged to using the simulated STM image as input. The model achieved an average F1 score of around 0.9 for each lattice type.

DL has also been used to improve the analysis of electron backscatter diffraction (EBSD) data, with Liu et al. 218 presenting one of the first DL-based solution for EBSD indexing capable of taking an EBSD image as input and predicting the three Euler angles representing the orientation that would have led to the given EBSD pattern. However, they considered the three Euler angles to be independent of each other, creating separate CNNs for each angle, although the three angles should be considered together. Jha et al. 219 built upon that work to train a single DL model to predict the three Euler angles in simulated EBSD patterns of polycrystalline Ni while directly minimizing the misorientation angle between the true and predicted orientations. When tested on experimental EBSD patterns, the model achieved 16% lower disorientation error than dictionary-based indexing. Similarly, Kaufman et al. trained a CNN to predict the corresponding space group for a given diffraction pattern 220 . This enables EBSD to be used for phase identification in samples where the existing phases are unknown, providing a faster or more cost-effective method of characterizing than X-ray or neutron diffraction. The results from these studies demonstrate the promise of applying DL to improve the performance and utility of EBSD experiments.

Recently, DL has also been to learn crystal plasticity using images of strain profiles as input 221 , 222 . The work in ref. 221 used domain knowledge integration in the form of two-point auto-correlation to enhance the predictive accuracy, while 222 applied residual learning to learn crystal plasticity at nanoscale. It used strain profiles of materials of varying sample widths ranging from 2 μm down to 62.5 nm obtained from discrete dislocation dynamics to build a deep residual network capable of identifying prior deformation history of the sample as low, medium, or high. Compared to the correlation function-based method (68.24% accuracy), the DL model was found to be significantly more accurate (92.48%) and also capable of predicting stress-strain curves of test samples. This work additionally used saliency maps to try to interpret the developed DL model.

Pixelwise learning

DL can also be applied to generate one or more predictions for every pixel in an image. This can provide more detailed information about the size, position, orientation, and morphology of features of interest in images. Thus, pixelwise learning has been a significant area of focus with many recent studies appearing in materials science literature.

Azimi et al. applied an ensemble of fully convolutional neural networks to segment martensite, tempered martensite, bainite, and pearlite in SEM images of carbon steels. Their model achieved 94% accuracy, demonstrating a significant improvement over previous efforts to automate the segmentation of different phases in SEM images. Decost, Francis, and Holm applied PixelNet to segment microstructural constituents in the UltraHigh Carbon Steel Database 223 , 224 . In contrast to fully convolutional neural networks, which encode and decode visual signals using a series of convolution layers, PixelNet constructs “hypercolumns”, or concatenations of feature representations corresponding to each pixel at different layers in a neural network. The hypercolumns are treated as individual feature vectors, which can then be classified using any typical classification approach, like a multilayer perceptron. This approach achieved phase segmentation precision and recall scores of 86.5% and 86.5%, respectively. Additionally, this approach was used to segment spheroidite particles in the matrix, achieving precision and recall scores of 91.1% and 91.1%, respectively.

Pixelwise DL has also been applied to automatically segment dislocations in Ni superalloys 210 . Dislocations are visually similar to \(\gamma -{\gamma }^{\prime}\) and dislocation in Ni superalloys. With limited training data, a single segmentation model could not distinguish between these features. To overcome this, a second model was trained to generate a coarse mask corresponding to the deformed region in the material. Overlaying this mask with predictions from the first model selects the dislocations, enabling them to be distinguished from \(\gamma -{\gamma }^{\prime}\) interfaces.

Stan, Thompson, and Voorhees applied Pixelwise DL to characterize dendritic growth from serial sectioning and synchrotron computed tomography data 225 . Both of these techniques generate large amounts of data, making manual analysis impractical. Conventional image processing approaches, utilizing thresholding, edge detectors, or other hand-crafted filters, cannot effectively deal with noise, contrast gradients, and other artifacts that are present in the data. Despite having a small training set of labeled images, SegNet automatically segmented these images with much higher performance.

Object/entity recognition, localization, and tracking

Object detection or localization is needed when individual instances of recognized objects in a given image need to be distinguished from each other. In cases where instances do not overlap each other by a significant amount, individual instances can be resolved through post-processing of semantic segmentation outputs. This technique has been applied extensively to detect individual atoms and defects in microstructural images.

Madsen et al. applied pixelwise DL to detect atoms in simulated atomic-resolution TEM images of graphene 226 . A neural network was trained to detect the presence of each atom as well as predict its column height. Pixelwise results are used as seeds for watershed segmentation to achieve instance-level detection. Analysis of the arrangement of the atoms led to the autonomous characterization of defects in the lattice structure of the material. Interestingly, despite being trained only on simulations, the model successfully detected atomic positions in experimental images.

Maksov et al. demonstrated atomistic defect recognition and tracking across sequences of atomic-resolution STEM images of WS 2 227 . The lattice structure and defects existing in the first frame were characterized through a physics-based approach utilizing Fourier transforms. The positions of atoms and defects in the first frame were used to train a segmentation model. Despite only using the first frame for training, the model successfully identified and tracked defects in the subsequent frames for each sequence, even when the lattice underwent significant deformation. Similarly, Yang et al. 228 used U-net architecture (as shown in Fig. 4 ) to detect vacancies and dopants in WSe 2 in STEM images with model accuracy of up to 98%. They classified the possible atomic sites based on experimental observations into five different types: tungsten, vanadium substituting for tungsten, selenium with no vacancy, mono-vacancy of selenium, and di-vacancy of selenium.

a Deep neural networks U-Net model constructed for quantification analysis of annular dark-field in the scanning transmission electron microscope (ADF-STEM) image of V-WSe 2 . b Examples of training dataset for deep learning of atom segmentation model for five different species. c Pixel-level accuracy of the atom segmentation model as a function of training epoch. d Measurement accuracy of the segmentation model compared with human-based measurements. Scale bars are 1 nm [Reprinted according to the terms of the CC-BY license ref. 228 ].

Roberts et al. developed DefectSegNet to automatically identify defects in transmission and STEM images of steel including dislocations, precipitates, and voids 229 . They provide detailed information on the model’s design, training, and evaluation. They also compare measurements generated from the model to manual measurements performed by several different human experts, demonstrating that the measurements generated by DL are quantitatively more accurate and consistent.

Kusche et al. applied DL to localize defects in panoramic SEM images of dual-phase steel 230 . Manual thresholding was applied to identify dark defects against the brighter matrix. Regions containing defects were classified via two neural networks. The first neural network distinguished between inclusions and ductile damage in the material. The second classified the type of ductile damage (i.e., notching, martensite cracking, etc.) Each defect was also segmented via a watershed algorithm to obtain detailed information on its size, position, and morphology.

Applying DL to localize defects and atomic structures is a popular area in materials science research. Thus, several other recent studies on these applications can be found in the literature 231 , 232 , 233 , 234 .

In the above examples pixelwise DL, or classification models are combined with image analysis to distinguish individual instances of detected objects. However, when several adjacent objects of the same class touch or overlap each other in the image, this approach will falsely detect them to be a single, larger object. In this case, DL models designed for the detection or instance segmentation can be used to resolve overlapping instances. In one such study, Cohn and Holm applied DL for instance-level segmentation of individual particles and satellites in dense powder images 235 . Segmenting each particle allows for computer vision to generate detailed size and morphology information which can be used to supplement experimental powder characterization for additive manufacturing. Additionally, overlaying the powder and satellite masks yielded the first method for quantifying the satellite content of powder samples, which cannot be measured experimentally.

Super-resolution imaging and auto-tuning experimental parameters

The studies listed so far focus on automating the analysis of existing data after it has been collected experimentally. However, DL can also be applied during experiments to improve the quality of the data itself. This can reduce the time for data collection or improve the amount of information captured in each image. Super-resolution and other DL techniques can also be applied in situ to autonomously adjust experimental parameters.

Recording high-resolution electron microscope images often require large dwell times, limiting the throughput of microscopy experiments. Additionally, during imaging, interactions between the electron beam and a microscopy sample can result in undesirable effects, including charging of non-conductive samples and damage to sensitive samples. Thus, there is interest in using DL to artificially increase the resolution of images without introducing these artifacts. One method of interest is applying generative adversarial networks (GANs) for this application.

De Haan et al. recorded SEM images of the same regions of interest in carbon samples containing gold nanoparticles at two resolutions 236 . Low-resolution images recorded were used as inputs to a GAN. The corresponding images with twice the resolution were used as the ground truth. After training the GAN reduced the number of undetected gaps between nanoparticles from 13.9 to 3.7%, indicating that super-resolution was successful. Thus, applying DL led to a four-fold reduction of the interaction time between the electron beam and the sample.

Ede and Beanland collected a dataset of STEM images of different samples 237 . Images were subsampled with spiral and ‘jittered’ grid masks to obtain partial images with resolutions reduced by a factor up to 100. A GAN was trained to reconstruct full images from their corresponding partial images. The results indicated that despite a significant reduction in the sampling area, this approach successfully reconstructed high-resolution images with relatively small errors.

DL has also been applied to automated tip conditioning for SPM experiments. Rashidi and Wolkow trained a model to detect artifacts in SPM measurements resulting from degradation in tip quality 238 . Using an ensemble of convolutional neural networks resulted in 99% accuracy. After detecting that a tip has degraded, the SPM was configured to automatically recondition the tip in situ until the network indicated that the atomic sharpness of the tip has been restored. Monitoring and reconditioning the tip is the most time and labor-intensive part of conducting SPM experiments. Thus, automating this process through DL can increase the throughput and decrease the cost of collecting data through SPM.

In addition to materials characterization, DL can be applied to autonomously adjust parameters during manufacturing. Scime et al. mounted a camera to multiple 3D printers 239 . Images of the build plate were recorded throughout the printing process. A dynamic segmentation convolutional neural network was trained to recognize defects such as recoater streaking, incomplete spreading, spatter, porosity, and others. The trained model achieved high performance and was transferable to multiple printers from three different methods of additive manufacturing. This work is the first step to enabling smart additive manufacturing machines that can correct defects and adjust parameters during printing.

There is also growing interest in establishing instruments and laboratories for autonomous experimentation. Eppel et al. trained multiple models to detect chemicals, materials, and transparent vessels in a chemistry lab setting 240 . This study provides a rigorous analysis of several different approaches for scene understanding. Models were trained to characterize laboratory scenes with different methods including semantic segmentation and instance segmentation, both with and without overlapping instances. The models successfully detected individual vessels and materials in a variety of settings. Finer-grained understanding of the contents of vessels, such as segmentation of individual phases in multi-phase systems, was limited, outlining the path for future work in this area. The results represent an important step towards realizing automated experimentation for laboratory-scale experiments.

Microstructure representation learning

Materials microstructure is often represented in the form of multi-phase high-dimensional 2D/3D images and thus can readily leverage image-based DL methods to learn robust, low-dimensional microstructure representations, which can subsequently be used for building predictive and generative models to learn forward and inverse structure-property linkages, which are typically studied across different length scales (multi-scale modeling). In this context, homogenization and localization refer to the transfer of information from lower length scales to higher length scales and vice-versa. DL using customized CNNs has been used both for homogenization, i.e., predicting the macroscale property of material given its microstructure information 221 , 241 , 242 , as well as for localization, i.e., predicting the strain distribution across a given microstructure for a loading condition 243 .

Transfer learning has also been widely used for analyzing materials microstructure images; methods for improving the use of transfer learning to materials science applications remain an area of active research. Goetz et al. investigated the use of unsupervised domain adaptation as an alternative to simply fine-tuning a pre-trained model 244 . In this technique a model is first trained on a labeled dataset in the source domain. Next, a discriminator model is used to train the model to generate domain-agnostic features. Compared to simple fine-tuning, unsupervised domain adaptation improved the performance of classification and segmentation neural networks on materials science datasets. However, it was determined that the highest performance was achieved when the source domain was more visually similar to the target (for example, using a different set of microstructural images instead of ImageNet.) This highlights the utility of establishing large, publicly available datasets of annotated images in materials science.

Kitaraha and Holm used the output of an intermediate layer of a pre-trained convolutional neural network as a feature representation for images of steel surface defects and Inconnel fracture surfaces 245 . Images were classified by defect type or fracture surface orientation using unsupervised DL. Even though no labeled data was used to train the neural network or the unsupervised classifier, the model found natural decision boundaries that achieved a classification performance of 98% and 88% for the defect classes and fracture surface orientations, respectively. Visualization of the representations through principal component analysis (PCA) and t-distributed stochastic neighborhood embedding (t-SNE) provided qualitative insights into the representations. Although the detailed physical interpretation of the representations is still a distant goal, this study provides tools for investigating patterns in visual signals contained in image-based datasets in materials science.

Larmuseau et al. investigated the use of triplet networks to obtain consistent representations for visually similar images of materials 246 . Triplet networks are trained with three images at a time. The first image, the reference, is classified by the network. The second image, called the positive, is another image with the same class label. The last image, called the negative, is an image from a separate class. During training the loss function includes errors in predicting the class of the reference image, the difference in representations of the reference and positive images, and the similarity in representations of the reference and negative images. This process allows the network to learn consistent representations for images in the same class while distinguishing images from different classes. The triple network outperformed an ordinary convolutional neural network trained for image classification on the same dataset.

In addition to investigating representations used to analyze existing images, DL can generate synthetic images of materials systems. Generative Adversarial Networks (GANs) are currently the predominant method for synthetic microstructure generation. GANs consist of a generator, which creates a synthetic microstructure image, and a discriminator, which attempts to predict if a given input image is real or synthetic. With careful application, GANs can be a powerful tool for microstructure representation learning and design.

Yang and Li et al. 247 , 248 developed a GAN-based model for learning a low-dimensional embedding of microstructures, which could then be easily sampled and used with the generator of the GAN model to generate realistic, statistically similar microstructure images, thus enabling microstructural materials design. The model was able to capture complex, nonlinear microstructure characteristics and learn the mapping between the latent design variables and microstructures. In order to close the loop, the method was combined with a Bayesian optimization approach to design microstructures with optimal optical absorption performance. The discovered microstructures were found to have up to 17% better property than randomly sampled microstructures. The unique architecture of their GAN model also facilitated generator scalability to generate arbitrary-sized microstructure images and discriminator transferability to build structure-property prediction models. Yang et al. 249 recently combined GANs with MDNs (mixture density networks) to enable inverse modeling in microstructural materials design, i.e., generate the microstructure for a given desired property.

Hsu et al. constructed a GAN to generate 3D synthetic solid oxide fuel cell microstructures 250 . These microstructures were compared to other synthetic microstructures generated by DREAM.3D as well as experimentally observed microstructures measured via sectioning and imaging with PFIB-SEM. Synthetic microstructures generated from the GAN were observed to qualitatively show better agreement to the experimental microstructures than the DREAM.3D microstructures, as evidenced by the more realistic phase connectivity and lower amount of agglomeration of solid phases. Additionally, a statistical analysis of various features such as volume fraction, particle size, and several other quantities demonstrated that the GAN microstructures were quantitatively more similar to the real microstructures than the DREAM.3D microstructures.

In a similar study, Chun et al. generated synthetic microstructures of high energy materials using a GAN 251 . Once again, a synthetic microstructure generated via GAN showed better qualitative visual similarity to an experimentally observed microstructure compared to a synthetic microstructure generated via a transfer learning approach, with sharper phase boundaries and fewer computational artifacts. Additionally, a statistical analysis of the void size, aspect ratio, and orientation distributions indicated that the GAN produced microstructures that were quantitatively more similar to real materials.

Applications of DL to microstructure representation learning can help researchers improve the performance of predictive models used for the applications listed above. Additionally, using generative models can generate more realistic simulated microstructures. This can help researchers develop more accurate models for predicting material properties and performance without needing to synthesize and process these materials, significantly increasing the throughput of materials selection and screening experiments.

Mesoscale modeling applications

In addition to image-based characterization, deep learning methods are increasingly used in mesoscale modeling. Dai et al. 252 trained a GNN successfully trained to predict magnetostriction in a wide range of synthetic polycrystalline systems with around 10% prediction error. The microstructure is represented by a graph where each node corresponds to a single grain, and the edges between nodes indicate an interface between neighboring grains. Five node features (3 Euler angles, volume, and the number of neighbors) were associated with each grain. The GNN outperformed other machine learning approaches for property prediction of polycrystalline materials by accounting for interactions between neighboring grains.

Similarly, Cohn and Holm present preliminary work applying GNNs to predict the occurrence of abnormal grain growth (AGG) in Monte Carlo simulations of microstructure evolution 253 . AGG appears to be stochastic, making it notoriously difficult to predict, control, and even observe experimentally in some materials. AGG has been reproduced in Monte Carlo simulations of material systems, but a model that can predict which initial microstructures will undergo AGG has not been established before. A dataset of Monte Carlo simulations was created using SPPARKS 254 , 255 . A microstructure GNN was trained to predict AGG in individual simulations, with 75% classification accuracy. In comparison, an image-based only achieved 60% accuracy. The GNN also provided physical insight to understanding AGG and indicated that only 2 neighborhood shells are needed to achieve the maximum performance achieved in the study. These early results motivate additional work on applying GNNs to predict the occurrence in both simulated and real materials during processing.

Natural language processing

Most of the existing knowledge in the materials domain is currently unavailable as structured information and only exists as unstructured text, tables, or images in various publications. There exists a great opportunity to use natural language processing (NLP) techniques to convert text to structured data or to directly learn and make inferences from the text information. However, as a relatively new field within materials science, many challenges remain unsolved in this domain, such as resolving dependencies between words and phrases across multiple sentences and paragraphs.

Datasets for NLP

Datasets relevant to natural language processing include peer-reviewed journal articles, articles published on preprint servers such as arXiv or ChemRxiv, patents, and online material such as Wikipedia. Unfortunately, accessing or parsing most such datasets remains difficult. Peer-reviewed journal articles are typically subject to copyright restrictions and thus difficult to obtain, especially in the large numbers required for machine learning. Many publishers now offer text and data mining (TDM) agreements that can be signed online, allowing at least a limited, restricted amount of work to be performed. However, gaining access to the full text of many publications still typically requires strict and dedicated agreements with each publisher. The major advantage of working with publishers is that they have often already converted the articles from a document format such as PDF into an easy-to-parse format such as HyperText Markup Language (HTML). In contrast, articles on preprint servers and patents are typically available with fewer restrictions, but are commonly available only as PDF files. It remains difficult to properly parse text from PDF files in a reliable manner, even when the text is embedded in the PDF. Therefore, new tools that can easily and automatically convert such content into well-structured HTML format with few residual errors would likely have a major impact on the field. Finally, online sources of information such as Wikipedia can serve as another type of data source. However, such online sources are often more difficult to verify in terms of accuracy and also do not contain as much domain-specific information as the research literature.

Software libraries for NLP

Applying NLP to a raw dataset involves multiple steps. These steps include retrieving the data, various forms of “pre-processing” (sentence and word tokenization, word stemming and lemmatization, featurization such as word vectors or part of speech tagging), and finally machine learning for information extraction (e.g., named entity recognition, entity-relationship modeling, question and answer, or others). Multiple software libraries exist to aid in materials NLP, as described in Table 5 . We note that although many of these steps can in theory be performed by general-purpose NLP libraries such as NLTK 256 , SpaCy 257 , or AllenNLP 258 , the specialized nature of chemistry and materials science text (including the presence of complex chemical formulas) often leads to errors. For example, researchers have developed specialized codes to perform preprocessing that better detect chemical formulas (and not split them into separate tokens or apply stemming/lemmatization to them) and scientific phrases and notation such as oxidation states or symbols for physical units.

Similarly, chemistry-specific codes for extracting entities are better at extracting the names of chemical elements (e.g., recognizing that “He” likely represents helium and not a male pronoun) and abbreviations for chemical formulas. Finally, word embeddings that convert words such as “manganese” into numerical vectors for further data mining are more informative when trained specifically on materials science text versus more generic texts, even when the latter datasets are larger 259 . Thus, domain-specific tools for NLP are required in nearly all aspects of the pipeline. The main exception is that the architecture of the specific neural network models used for information extraction (e.g., LSTM, BERT, or architectures used to generate word embeddings such as word2vec or GloVe) are typically not modified specifically for the materials domain. Thus, much of the materials and chemistry-centric work currently regards data retrieval and appropriate preprocessing. A longer discussion of this topic, with specific examples, can be found in refs. 260 , 261 .

NLP methods for materials have been applied for information extraction and search (particularly as applied to synthesis prediction) as well as materials discovery. As the domain is rapidly growing, we suggest dedicated reviews on this topic by Olivetti et al. 261 and Kononova et al. 260 for more information.

One of the major uses of NLP methods is to extract datasets from the text in published studies. Conventionally, such datasets required manual entry of datasets by researchers combing the literature, a laborious and time-consuming process. Recently, software tools such as ChemDataExtractor 262 and other methods 263 based on more conventional machine learning and rule-based approaches have enabled automated or semi-automated extraction of datasets such as Curie and Néel magnetic phase transition temperatures 264 , battery properties 265 , UV-vis spectra 266 , and surface and pore characteristics of metal-organic frameworks 267 . In the past few years, DL approaches such as LSTMs and transformer-based models have been employed to extract various categories of information 268 , and in particular materials synthesis information 269 , 270 , 271 from text sources. Such data have been used to predict synthesis maps for titania nanotubes 272 , various binary and ternary oxides 273 , and perovskites 274 .

Databases based on natural language processing have also been used to train machine learning models to identify materials with useful functional properties, such as the recent discovery of the large magnetocaloric properties of HoBe 2 275 . Similarly, Cooper et al. 276 demonstrated a “design to device approach” for designing dye-sensitized solar cells that are co-sensitized with two dyes 276 . This study used automated text mining to compile a list of candidate dyes for the application along with measured properties such as maximum absorption wavelengths and extinction coefficients. The resulting list of 9431 dyes extracted from the literature was downselected to 309 candidates using various criteria such as molecular structure and ability to absorb in the solar spectrum. These candidates were evaluated for suitable combinations for co-sensitization, yielding 33 dyes that were further downselected using density functional theory calculations and experimental constraints. The resulting 5 dyes were evaluated experimentally, both individually and in combinations, resulting in a combination of dyes that not only outperformed any of the individual dyes but demonstrated performance comparable to existing standard material. This study demonstrates the possibility of using literature-based extraction to identify materials candidates for new applications from the vast body of published work, which may have never tested those materials for the desired application.

It is even possible that natural language processing can directly make materials predictions without intermediary models. In a study reported by Tshitoyan et al. 259 (as shown in Fig. 5 ), word embeddings (i.e., numerical vectors representing distinct words) trained on materials science literature could directly predict materials applications through a simple dot product between the trained embedding for a composition word (such as PbTe) and an application words (such as thermoelectrics). The researchers demonstrated that such an approach, if applied in the past using historical data, may have subsequently predicted many recently reported thermoelectric materials; they also presented a list of potentially interesting thermoelectric compositions using the known literature at the time. Since then, several of these predictions have been tested either computationally 277 , 278 , 279 , 280 , 281 , 282 or experimentally 283 as potential thermoelectrics. Such approaches have recently been applied to search for understudied areas of metallocene catalysis 284 , although challenges still remain in such direct approaches to materials prediction.

a Network for training word embeddings for natural language processing application. A one-hot encoded vector at left represents each distinct word in the corpus; the role of a hidden layer is to predict the probability of neighboring words in the corpus. This network structure trains a relatively small hidden layer of 100–200 neurons to contain information on the context of words in the entire corpus, with the result that similar words end up with similar hidden layer weights (word embeddings). Such word embeddings can transform wordsin text form into numerical vectors that may be useful for a variety of applications. b projection of word embeddings for various materials science words, as trained on a corpus scientific abstracts, into two dimensions using principle components analysis. Without any explicit training, the word embeddings naturally preserve relationships between chemical formulas, their common oxides, and their ground state structures. [Reprinted according to the terms of the CC-BY license ref. 259 ].

Uncertainty quantification

Uncertainty quantification (UQ) is an essential step in evaluating the robustness of DL. Specifically, DL models have been criticized for lack of robustness, interpretability, and reliability and the addition of carefully quantified uncertainties would go a long way towards addressing such shortcomings. While most of the focus in the DL field currently goes into developing new algorithms or training networks to high accuracy, there is increasing attention to UQ, as exemplified by the detailed review of Abdar et al. 285 . However, determining the uncertainty associated with DL predictions is still challenging and far from a completely solved problem.

The main drawback to estimating UQ when performing DL is the fact that most of the currently available UQ implementations do not work for arbitrary, off-the-shelf models, without retraining or redesigning. Bayesian NNs are the exception; however, they require significant modifications to the training procedure, are computationally expensive compared to non-Bayesian NNs, and become increasingly inefficient the larger the datasize gets. A considerable fraction of the current research in DL UQ focuses exactly on such an issue: how to evaluate uncertainty without requiring computationally expensive retraining or DL code modifications. An example of such an effort is the work of Mi et al. 286 , where three scalable methods are explored, to evaluate the variance of output from trained NN, without requiring any amount of retraining. Another example is Teye, Azizpour, and Smith’s exploration of the use of batch normalization as a way to approximate inference in Bayesian models 287 .

Before reviewing the most common methods used to evaluate uncertainty in DL, let us briefly point out key reasons to add UQ to DL modeling. Reaching high accuracy when training DL models implicitly assume the availability of a sufficiently large and diverse training dataset. Unfortunately, this rarely occurs in material discovery applications 288 . ML/DL models are prone to perform poorly on extrapolation 289 . It is also extremely difficult for ML/DL models to recognize ambiguous samples 290 . In general, determining the amount of data necessary to train a DL to achieve the required accuracy is a challenging problem. Careful evaluation of the uncertainty associated with DL predictions would not only increase reliability in predicted results but would also provide guidance on estimating the needed training dataset size as well as suggesting what new data should be added to reach the target accuracy (uncertainty-guided decision). Zhang, Kailkhura, and Han’s work emphasizes how including a UQ-motivated reject option into the DL model substantially improves the performance of the remaining material data 288 . Such a reject option is associated with the detection of out-of-distribution samples, which is only possible through UQ analysis of the predicted results.

Two different uncertainty types are associated with each ML prediction: epistemic uncertainty and aleatory uncertainty. Epistemic uncertainty is related to insufficient training data in part of the input domain. As mentioned above, while DL is very effective at interpolation tasks, they can have more difficulty in extrapolation. Therefore, it is vital to quantify the lack of accuracy due to localized, insufficient training data. The aleatory uncertainty, instead, is related to parameters not included in the model. It relates to the possibility of training on data that our DL perceives as very similar but that are associated with different outputs because of missing features in the model. Ideally, we would like UQ methodologies to distinguish and quantify both types of uncertainties separately.

The most common approaches to evaluate uncertainty using DL are Dropout methods, Deep Ensemble methods, Quantile regression, and Gaussian Processes. Dropout methods are commonly used to avoid overfitting. In this type of approach, network nodes are disabled randomly during training, resulting in the evaluation of a different subset of the network at each training step. When a similar randomization procedure is also applied to the prediction procedure, the methodology becomes Monte-Carlo dropout 291 . Repeating such randomization multiple times produces a distribution over the outputs, from which mean and variance are determined for each prediction. Another example of using a dropout approach to approximate Bayesian inference in deep Gaussian processes is the work of Gal and Ghahramani 292 .

Deep ensemble methodologies 293 , 294 , 295 , 296 combine deep learning modelling with ensemble learning. Ensemble methods utilize multiple models and different random initializations to improve predictability. Because of the multiple predictions, statistical distributions of the outputs are generated. Combining such results into a Gaussian distribution, confidence intervals are obtained through variance evaluation. Such a multi-model strategy allows the evaluation of aleatory uncertainty when sufficient training data are provided. For areas without sufficient data, the predicted mean and variance will not be accurate, but the expectation is that a very large variance should be estimated, clearly indicating non-trustable predictions. Monte-Carlo Dropout and Deep Ensembles approaches can be combined to further improve confidence in the predicted outputs.

Quantile regression can be utilized with DL 297 . In this approach, the loss function is used in a way that allows to predict for the chosen quantile a (between 0 and 1). A choice of a  = 0.5 corresponds to evaluating the Mean Absolute Error (MAE) and predicting the median of the distribution. Predicting for two more quantile values (amin and amax) determines confidence intervals of width amax − amin. For instance, predicting for amin = 0.1 and amax = 0.8 produces confidence intervals covering 70% of the population. The largest drawback of using quantile to estimate prediction intervals is the need to run the model three times, one for each quantile needed. However, a recent implementation in TensorFlow allows to simultaneously obtain multiple quantiles in one run.

Lastly, Gaussian Processes (GP) can be used within a DL approach as well and have the side benefit of providing UQ information at no extra cost. Gaussian processes are a family of infinite-dimensional multivariate Gaussian distributions completely specified by a mean function and a flexible kernel function (prior distribution). By optimizing such functions to fit the training data, the posterior distribution is determined, which is later used to predict outputs for inputs not included in the training set. Because the prior is a Gaussian process, the posterior distribution is Gaussian as well 298 , thus providing mean and variance information for each predicted data. However, in practice standard kernels under-perform 299 . In 2016, Wilson et al. 300 suggested processing inputs through a neural network prior to a Gaussian process model. This procedure could extract high-level patterns and features, but required careful design and optimization. In general, Deep Gaussian processes improve the performance of Gaussian processes by mapping the inputs through multiple Gaussian process ‘layers’. Several groups have followed this avenue and further perfected such an approach (ref. 299 and references within). A common drawback of Bayesian methods is a prohibitive computational cost if dealing with large datasets 292 .

Limitations and challenges

Although DL methods have various fascinating opportunities for materials design, they have several limitations and there is much room to improve. Reliability and quality assessment of datasets used in DL tasks are challenging because there is either a lack of ground truth data, or there are not enough metrics for global comparison, or datasets using similar or identical set-ups may not be reproducible 301 . This poses an important challenge in relying upon DL-based prediction.

Material representations based on chemical formula alone by definition do not consider structure, which on the one hand makes them more amenable to work for new compounds for which structure information may not be available, but on the other hand, makes it impossible for them to capture phenomena such as phase transitions. Properties of materials depend sensitively on structure to the extent that their properties can be quite opposite depending on the atomic arrangement, like a diamond (hard, wide-band-gap insulator) and graphite (soft, semi-metal). It is thus not a surprise that chemical formula-based methods may not be adequate in some cases 159 .

Atomistic graph-based predictions, although considered a full atomistic description, are tested on bulk materials only and not for defective systems or for multi-dimensional phases of space exploration such as using genetic algorithms. In general, this underscores that the input features must be predictive for the output labels and not be missing some key information. Although atomistic graph neural network models such as atomistic line graph neural network (ALIGNN) have achieved remarkable accuracy compared to previous atomistic based models, the model errors still need to be further brought down to reach something resembling deep learning ‘chemical-accuracies.’

In terms of images and spectra, the experimental data are too noisy most of the time and require much manipulation before applying DL. In contrast, theory-based simulated data represent an alternate path forward but may not capture realistic scenarios such as the presence of structured noise 217 .

Uncertainty quantification for deep learning for materials science is important, yet only a few works have been published in this field. To alleviate the black-box 38 nature of the DL methods, a package such as GNNExplainer 302 has been tried in the context of the material. Such attempts at greater interpretability will be important moving forward to gain the trust of the materials community.

While training-validation-test split strategies were primarily designed in DL for image classification tasks with a certain number of classes, the same for regression models in materials science may not be the best approach. This is because it is possible that during the training the model is seeing a material very similar to the test set material and in reality it is difficult to generalize the model. Best practices need to be developed for data split, normalization, and augmentation to avoid such issues 289 .

Finally, we note an important technological challenge is to make a closed-loop autonomous materials design and synthesis process 303 , 304 that can include both machine learning and experimental components in a self-driving laboratory 305 . For an overview of early proof of principle attempts see 306 . For example, in an autonomous synthesis experiment the oxidation state of copper (and therefore the oxide phase) was varied in a sample of copper oxide by automatically flowing more oxidizing or more reducing gas over the sample and monitoring the charge state of the copper using XANES. An algorithmic decision policy was then used to automatically change the gas composition for a subsequent experiment based on the prior experiments, with no human in the loop, in such a way as to autonomously move towards a target copper oxidation state 307 . This simple proof of principle experiment provides just a glimpse of what is possible moving forward.

Data availability

The data from new figures are available on reasonable request from the corresponding author. Data from other publishers are not available from the corresponding author of this work but may be available by reaching the corresponding author of the cited work.

Code availability

Software packages mentioned in the article (whichever made available by the authors) can be found at https://github.com/deepmaterials/dlmatreview . Software for other packages can be obtained by reaching the corresponding author of the cited work.

Callister, W. D. et al. Materials Science and Engineering: An Introduction (Wiley, 2021).

Saito, T. Computational Materials Design, Vol. 34 (Springer Science & Business Media, 2013).

Choudhary, K. et al. The joint automated repository for various integrated simulations (jarvis) for data-driven materials design. npj Comput. Mater. 6 , 1–13 (2020).

Article   Google Scholar  

Kirklin, S. et al. The open quantum materials database (oqmd): assessing the accuracy of dft formation energies. npj Comput. Mater. 1 , 1–15 (2015).

Jain, A. et al. Commentary: The materials project: A materials genome approach to accelerating materials innovation. APL Mater. 1 , 011002 (2013).

Curtarolo, S. et al. Aflow: An automatic framework for high-throughput materials discovery. Comput. Mater. Sci. 58 , 218–226 (2012).

Article   CAS   Google Scholar  

Ramakrishnan, R., Dral, P. O., Rupp, M. & Von Lilienfeld, O. A. Quantum chemistry structures and properties of 134 kilo molecules. Sci. Data 1 , 1–7 (2014).

Draxl, C. & Scheffler, M. Nomad: The fair concept for big data-driven materials science. MRS Bull. 43 , 676–682 (2018).

Wang, R., Fang, X., Lu, Y., Yang, C.-Y. & Wang, S. The pdbbind database: methodologies and updates. J. Med. Chem. 48 , 4111–4119 (2005).

Zakutayev, A. et al. An open experimental database for exploring inorganic materials. Sci. Data 5 , 1–12 (2018).

de Pablo, J. J. et al. New frontiers for the materials genome initiative. npj Comput. Mater. 5 , 1–23 (2019).

Wilkinson, M. D. et al. The fair guiding principles for sci. data management and stewardship. Sci. Data 3 , 1–9 (2016).

Friedman, J. et al. The Elements of Statistical Learning, Vol. 1 (Springer series in statistics New York, 2001).

Agrawal, A. & Choudhary, A. Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science. APL Mater. 4 , 053208 (2016).

Vasudevan, R. K. et al. Materials science in the artificial intelligence age: high-throughput library generation, machine learning, and a pathway from correlations to the underpinning physics. MRS Commun. 9 , 821–838 (2019).

Schmidt, J., Marques, M. R., Botti, S. & Marques, M. A. Recent advances and applications of machine learning in solid-state materials science. npj Comput. Mater. 5 , 1–36 (2019).

Butler, K. T., Davies, D. W., Cartwright, H., Isayev, O. & Walsh, A. Machine learning for molecular and materials science. Nature 559 , 547–555 (2018).

Xu, Y. et al. Deep dive into machine learning models for protein engineering. J. Chem. Inf. Model. 60 , 2773–2790 (2020).

Schleder, G. R., Padilha, A. C., Acosta, C. M., Costa, M. & Fazzio, A. From dft to machine learning: recent approaches to materials science–a review. J. Phys. Mater. 2 , 032001 (2019).

Agrawal, A. & Choudhary, A. Deep materials informatics: applications of deep learning in materials science. MRS Commun. 9 , 779–792 (2019).

Goodfellow, I., Bengio, Y. & Courville, A. Deep Learning (MIT Press, 2016).

LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521 , 436–444 (2015).

McCulloch, W. S. & Pitts, W. A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophys. 5 , 115–133 (1943).

Rosenblatt, F. The perceptron: A probabilistic model for information storage and organization in the brain. Psychol. Rev. 65 , 386–408 (1958).

Gibney, E. Google ai algorithm masters ancient game of go. Nat. News 529 , 445 (2016).

Ramos, S., Gehrig, S., Pinggera, P., Franke, U. & Rother, C. Detecting unexpected obstacles for self-driving cars: Fusing deep learning and geometric modeling. in 2017 IEEE Intelligent Vehicles Symposium (IV) , 1025–1032 (IEEE, 2017).

Buduma, N. & Locascio, N. Fundamentals of deep learning: Designing next-generation machine intelligence algorithms (O’Reilly Media, Inc., O’Reilly, 2017).

Kearnes, S., McCloskey, K., Berndl, M., Pande, V. & Riley, P. Molecular graph convolutions: moving beyond fingerprints. J. Computer Aided Mol. Des. 30 , 595–608 (2016).

Albrecht, T., Slabaugh, G., Alonso, E. & Al-Arif, S. M. R. Deep learning for single-molecule science. Nanotechnology 28 , 423001 (2017).

Ge, M., Su, F., Zhao, Z. & Su, D. Deep learning analysis on microscopic imaging in materials science. Mater. Today Nano 11 , 100087 (2020).

Agrawal, A., Gopalakrishnan, K. & Choudhary, A. In Handbook on Big Data and Machine Learning in the Physical Sciences: Volume 1. Big Data Methods in Experimental Materials Discovery World Scientific Series on Emerging Technologies, 205–230 (“World Scientific, 2020).

Erdmann, M., Glombitza, J., Kasieczka, G. & Klemradt, U. Deep Learning for Physics Research (World Scientific, 2021).

Chen, C., Ye, W., Zuo, Y., Zheng, C. & Ong, S. P. Graph networks as a universal machine learning framework for molecules and crystals. Chem. Mater. 31 , 3564–3572 (2019).

Jha, D. et al. Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning. Nat. Commun . 10 , 1–12 (2019).

Cubuk, E. D., Sendek, A. D. & Reed, E. J. Screening billions of candidates for solid lithium-ion conductors: a transfer learning approach for small data. J. Chem. Phys. 150 , 214701 (2019).

Chen, C., Zuo, Y., Ye, W., Li, X. & Ong, S. P. Learning properties of ordered and disordered materials from multi-fidelity data. Nat. Comput. Sci. 1 , 46–53 (2021).

Artrith, N. et al. Best practices in machine learning for chemistry. Nat. Chem. 13 , 505–508 (2021).

Holm, E. A. In defense of the black box. Science 364 , 26–27 (2019).

Mueller, T., Kusne, A. G. & Ramprasad, R. Machine learning in materials science: Recent progress and emerging applications. Rev. Comput. Chem. 29 , 186–273 (2016).

CAS   Google Scholar  

Wei, J. et al. Machine learning in materials science. InfoMat 1 , 338–358 (2019).

Liu, Y. et al. Machine learning in materials genome initiative: a review. J. Mater. Sci. Technol. 57 , 113–122 (2020).

Wang, A. Y.-T. et al. Machine learning for materials scientists: an introductory guide toward best practices. Chem. Mater. 32 , 4954–4965 (2020).

Morgan, D. & Jacobs, R. Opportunities and challenges for machine learning in materials science. Annu. Rev. Mater. Res. 50 , 71–103 (2020).

Himanen, L., Geurts, A., Foster, A. S. & Rinke, P. Data-driven materials science: status, challenges, and perspectives. Adv. Sci. 6 , 1900808 (2019).

Rajan, K. Informatics for materials science and engineering: data-driven discovery for accelerated experimentation and application (Butterworth-Heinemann, 2013).

Montáns, F. J., Chinesta, F., Gómez-Bombarelli, R. & Kutz, J. N. Data-driven modeling and learning in science and engineering. Comptes Rendus Mécanique 347 , 845–855 (2019).

Aykol, M. et al. The materials research platform: defining the requirements from user stories. Matter 1 , 1433–1438 (2019).

Stanev, V., Choudhary, K., Kusne, A. G., Paglione, J. & Takeuchi, I. Artificial intelligence for search and discovery of quantum materials. Commun. Mater. 2 , 1–11 (2021).

Chen, C. et al. A critical review of machine learning of energy materials. Adv. Energy Mater. 10 , 1903242 (2020).

Cybenko, G. Approximation by superpositions of a sigmoidal function. Math. Control Signals Syst. 2 , 303–314 (1989).

Kidger, P. & Lyons, T. Universal approximation with deep narrow networks . in Conference on learning theory , 2306–2327 (PMLR, 2020).

Lin, H. W., Tegmark, M. & Rolnick, D. Why does deep and cheap learning work so well? J. Stat. Phys. 168 , 1223–1247 (2017).

Minsky, M. & Papert, S. A. Perceptrons: An introduction to computational geometry (MIT press, 2017).

Paszke, A. et al. Pytorch: An imperative style, high-performance deep learning library. Adv. Neural Inf. Process. Syst. 32 , 8026–8037 (2019).

Google Scholar  

Abadi et al., TensorFlow: A system for large-scale machine learning. arXiv:1605.08695, Preprint at https://arxiv.org/abs/1605.08695 (2006).

Chen, T. et al. Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems. arXiv . https://arxiv.org/abs/1512.01274 (2015).

Nwankpa, C., Ijomah, W., Gachagan, A. & Marshall, S. Activation functions: comparison of trends in practice and research for deep learning. arXiv . https://arxiv.org/abs/1811.03378 (2018).

Baydin, A. G., Pearlmutter, B. A., Radul, A. A. & Siskind, J. M. Automatic differentiation in machine learning: a survey. J. Machine Learn. Res. 18 , 1–43 (2018).

Hinton, G. E., Srivastava, N., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. R. Improving neural networks by preventing co-adaptation of feature detectors. arXiv. https://arxiv.org/abs/1207.0580 (2012).

Breiman, L. Bagging predictors. Machine Learn. 24 , 123–140 (1996).

LeCun, Y. et al. The Handbook of Brain Theory and Neural Networks vol. 3361 (MIT press Cambridge, MA, USA 1995).

Wilson, R. J. Introduction to Graph Theory (Pearson Education India, 1979).

West, D. B. et al. Introduction to Graph Theory Vol. 2 (Prentice hall Upper Saddle River, 2001).

Wang, M. et al. Deep graph library: A graph-centric, highly-performant package for graph neural networks. arXiv . https://arxiv.org/abs/1909.01315 (2019).

Choudhary, K. & DeCost, B. Atomistic line graph neural network for improved materials property predictions. npj Comput. Mater. 7 , 1–8 (2021).

Li, M. et al. Dgl-lifesci: An open-source toolkit for deep learning on graphs in life science. arXiv . https://arxiv.org/abs/2106.14232 (2021).

Xie, T. & Grossman, J. C. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties. Phys. Rev. Lett. 120 , 145301 (2018).

Klicpera, J., Groß, J. & Günnemann, S. Directional message passing for molecular graphs. arXiv . https://arxiv.org/abs/2003.03123 (2020).

Schutt, K. et al. Schnetpack: A deep learning toolbox for atomistic systems. J. Chem. Theory Comput. 15 , 448–455 (2018).

Kipf, T. N. & Welling, M. Semi-supervised classification with graph convolutional networks. arXiv . https://arxiv.org/abs/1609.02907 (2016).

Veličković, P. et al. Graph attention networks. arXiv . https://arxiv.org/abs/1710.10903 (2017).

Schlichtkrull, M. et al. Modeling relational data with graph convolutional networks. arXiv. https://arxiv.org/abs/1703.06103 (2017).

Song, L., Zhang, Y., Wang, Z. & Gildea, D. A graph-to-sequence model for AMR-to-text generation . In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , 1616–1626 (Association for Computational Linguistics, 2018).

Xu, K., Hu, W., Leskovec, J. & Jegelka, S. How powerful are graph neural networks? arXiv . https://arxiv.org/abs/1810.00826 (2018).

Chen, Z., Li, X. & Bruna, J. Supervised community detection with line graph neural networks. arXiv . https://arxiv.org/abs/1705.08415 (2017).

Jing, Y., Bian, Y., Hu, Z., Wang, L. & Xie, X.-Q. S. Deep learning for drug design: an artificial intelligence paradigm for drug discovery in the big data era. AAPS J. 20 , 1–10 (2018).

Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv. https://arxiv.org/abs/1810.04805 (2018).

De Cao, N. & Kipf, T. Molgan: An implicit generative model for small molecular graphs. arXiv . https://arxiv.org/abs/1805.11973 (2018).

Pereira, T., Abbasi, M., Ribeiro, B. & Arrais, J. P. Diversity oriented deep reinforcement learning for targeted molecule generation. J. Cheminformatics 13 , 1–17 (2021).

Baker, N. et al. Workshop report on basic research needs for scientific machine learning: core technologies for artificial intelligence. Tech. Rep . https://doi.org/10.2172/1478744 . (2019).

Chan, H. et al. Rapid 3d nanoscale coherent imaging via physics-aware deep learning. Appl. Phys. Rev. 8 , 021407 (2021).

Pun, G. P., Batra, R., Ramprasad, R. & Mishin, Y. Physically informed artificial neural networks for atomistic modeling of materials. Nat. Commun. 10 , 1–10 (2019).

Onken, D. et al. A neural network approach for high-dimensional optimal control. arXiv. https://arxiv.org/abs/2104.03270 (2021).

Zunger, A. Inverse design in search of materials with target functionalities. Nat. Rev. Chem. 2 , 1–16 (2018).

Chen, L., Zhang, W., Nie, Z., Li, S. & Pan, F. Generative models for inverse design of inorganic solid materials. J. Mater. Inform. 1 , 4 (2021).

Cranmer, M. et al. Discovering symbolic models from deep learning with inductive biases. arXiv . https://arxiv.org/abs/2006.11287 (2020).

Rupp, M., Tkatchenko, A., Müller, K.-R. & Von Lilienfeld, O. A. Fast and accurate modeling of molecular atomization energies with machine learning. Phys. Rev. Lett. 108 , 058301 (2012).

Bartók, A. P., Kondor, R. & Csányi, G. On representing chemical environments. Phys. Rev. B 87 , 184115 (2013).

Faber, F. A. et al. Prediction errors of molecular machine learning models lower than hybrid dft error. J. Chem. Theory Comput. 13 , 5255–5264 (2017).

Choudhary, K., DeCost, B. & Tavazza, F. Machine learning with force-field-inspired descriptors for materials: Fast screening and mapping energy landscape. Phys. Rev. Mater. 2 , 083801 (2018).

Choudhary, K., Garrity, K. F., Ghimire, N. J., Anand, N. & Tavazza, F. High-throughput search for magnetic topological materials using spin-orbit spillage, machine learning, and experiments. Phys. Rev. B 103 , 155131 (2021).

Choudhary, K., Garrity, K. F. & Tavazza, F. Data-driven discovery of 3d and 2d thermoelectric materials. J. Phys. Condens. Matter 32 , 475501 (2020).

Ward, L. et al. Including crystal structure attributes in machine learning models of formation energies via voronoi tessellations. Phys. Rev. B 96 , 024104 (2017).

Isayev, O. et al. Universal fragment descriptors for predicting properties of inorganic crystals. Nat. Commun. 8 , 1–12 (2017).

Liu, C.-H., Tao, Y., Hsu, D., Du, Q. & Billinge, S. J. Using a machine learning approach to determine the space group of a structure from the atomic pair distribution function. Acta Crystallogr. Sec. A 75 , 633–643 (2019).

Smith, J. S., Isayev, O. & Roitberg, A. E. Ani-1: an extensible neural network potential with dft accuracy at force field computational cost. Chem. Sci. 8 , 3192–3203 (2017).

Behler, J. Atom-centered symmetry functions for constructing high-dimensional neural network potentials. J. Chem. Phys. 134 , 074106 (2011).

Behler, J. & Parrinello, M. Generalized neural-network representation of high-dimensional potential-energy surfaces. Phys. Rev. Lett. 98 , 146401 (2007).

Ko, T. W., Finkler, J. A., Goedecker, S. & Behler, J. A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer. Nat. Commun. 12 , 398 (2021).

Weinreich, J., Romer, A., Paleico, M. L. & Behler, J. Properties of alpha-brass nanoparticles. 1. neural network potential energy surface. J. Phys. Chem C 124 , 12682–12695 (2020).

Wang, H., Zhang, L., Han, J. & E, W. Deepmd-kit: A deep learning package for many-body potential energy representation and molecular dynamics. Computer Phys. Commun. 228 , 178–184 (2018).

Eshet, H., Khaliullin, R. Z., Kühne, T. D., Behler, J. & Parrinello, M. Ab initio quality neural-network potential for sodium. Phys. Rev. B 81 , 184107 (2010).

Khaliullin, R. Z., Eshet, H., Kühne, T. D., Behler, J. & Parrinello, M. Graphite-diamond phase coexistence study employing a neural-network mapping of the ab initio potential energy surface. Phys. Rev. B 81 , 100103 (2010).

Artrith, N. & Urban, A. An implementation of artificial neural-network potentials for atomistic materials simulations: Performance for tio2. Comput. Mater. Sci. 114 , 135–150 (2016).

Park, C. W. et al. Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture. npj Comput. Mater. 7 , 1–9 (2021).

Chmiela, S., Sauceda, H. E., Müller, K.-R. & Tkatchenko, A. Towards exact molecular dynamics simulations with machine-learned force fields. Nat. Commun. 9 , 1–10 (2018).

Xue, L.-Y. et al. Reaxff-mpnn machine learning potential: a combination of reactive force field and message passing neural networks. Phys. Chem. Chem. Phys. 23 , 19457–19464 (2021).

Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O. & Dahl, G. E. Neural message passing for quantum chemistry. arXiv . https://arxiv.org/abs/1704.01212 (2017).

Zitnick, C. L. et al. An introduction to electrocatalyst design using machine learning for renewable energy storage. arXiv. https://arxiv.org/abs/2010.09435 (2020).

McNutt, A. T. et al. Gnina 1 molecular docking with deep learning. J. Cheminformatics 13 , 1–20 (2021).

Jin, W., Barzilay, R. & Jaakkola, T. Junction tree variational autoencoder for molecular graph generation. in International conference on machine learning , 2323–2332 (PMLR, 2018).

Olivecrona, M., Blaschke, T., Engkvist, O. & Chen, H. Molecular de-novo design through deep reinforcement learning. J. Cheminformatics 9 , 1–14 (2017).

You, J., Liu, B., Ying, R., Pande, V. & Leskovec, J. Graph convolutional policy network for goal-directed molecular graph generation. arXiv. https://arxiv.org/abs/1806.02473 (2018).

Putin, E. et al. Reinforced adversarial neural computer for de novo molecular design. J. Chem. Inf. Model. 58 , 1194–1204 (2018).

Sanchez-Lengeling, B., Outeiral, C., Guimaraes, G. L. & Aspuru-Guzik, A. Optimizing distributions over molecular space. an objective-reinforced generative adversarial network for inverse-design chemistry (organic). ChemRxiv https://doi.org/10.26434/chemrxiv.5309668.v3 (2017).

Nouira, A., Sokolovska, N. & Crivello, J.-C. Crystalgan: learning to discover crystallographic structures with generative adversarial networks. arXiv. https://arxiv.org/abs/1810.11203 (2018).

Long, T. et al. Constrained crystals deep convolutional generative adversarial network for the inverse design of crystal structures. npj Comput. Mater. 7 , 66 (2021).

Noh, J. et al. Inverse design of solid-state materials via a continuous representation. Matter 1 , 1370–1384 (2019).

Kim, S., Noh, J., Gu, G. H., Aspuru-Guzik, A. & Jung, Y. Generative adversarial networks for crystal structure prediction. ACS Central Sci. 6 , 1412–1420 (2020).

Long, T. et al. Inverse design of crystal structures for multicomponent systems. arXiv. https://arxiv.org/abs/2104.08040 (2021).

Xie, T. & Grossman, J. C. Hierarchical visualization of materials space with graph convolutional neural networks. J. Chem. Phys. 149 , 174111 (2018).

Park, C. W. & Wolverton, C. Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery. Phys. Rev. Mater. 4 , 063801 (2020).

Laugier, L. et al. Predicting thermoelectric properties from crystal graphs and material descriptors-first application for functional materials. arXiv. https://arxiv.org/abs/1811.06219 (2018).

Rosen, A. S. et al. Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery. Matter 4 , 1578–1597 (2021).

Lusci, A., Pollastri, G. & Baldi, P. Deep architectures and deep learning in chemoinformatics: the prediction of aqueous solubility for drug-like molecules. J. Chem. Inf. Model. 53 , 1563–1575 (2013).

Xu, Y. et al. Deep learning for drug-induced liver injury. J. Chem. Inf. Model. 55 , 2085–2093 (2015).

Jain, A. & Bligaard, T. Atomic-position independent descriptor for machine learning of material properties. Phys. Rev. B 98 , 214112 (2018).

Goodall, R. E., Parackal, A. S., Faber, F. A., Armiento, R. & Lee, A. A. Rapid discovery of novel materials by coordinate-free coarse graining. arXiv . https://arxiv.org/abs/2106.11132 (2021).

Zuo, Y. et al. Accelerating Materials Discovery with Bayesian Optimization and Graph Deep Learning. arXiv . https://arxiv.org/abs/2104.10242 (2021).

Lin, T.-S. et al. Bigsmiles: a structurally-based line notation for describing macromolecules. ACS Central Sci. 5 , 1523–1531 (2019).

Tyagi, A. et al. Cancerppd: a database of anticancer peptides and proteins. Nucleic Acids Res. 43 , D837–D843 (2015).

Krenn, M., Häse, F., Nigam, A., Friederich, P. & Aspuru-Guzik, A. Self-referencing embedded strings (selfies): a 100% robust molecular string representation. Machine Learn. Sci. Technol. 1 , 045024 (2020).

Lim, J., Ryu, S., Kim, J. W. & Kim, W. Y. Molecular generative model based on conditional variational autoencoder for de novo molecular design. J. Cheminformatics 10 , 1–9 (2018).

Krasnov, L., Khokhlov, I., Fedorov, M. V. & Sosnin, S. Transformer-based artificial neural networks for the conversion between chemical notations. Sci. Rep. 11 , 1–10 (2021).

Irwin, J. J., Sterling, T., Mysinger, M. M., Bolstad, E. S. & Coleman, R. G. Zinc: a free tool to discover chemistry for biology. J. Chem. Inf. Model. 52 , 1757–1768 (2012).

Dix, D. J. et al. The toxcast program for prioritizing toxicity testing of environmental chemicals. Toxicol. Sci. 95 , 5–12 (2007).

Kim, S. et al. Pubchem 2019 update: improved access to chemical data. Nucleic Acids Res. 47 , D1102–D1109 (2019).

Hirohara, M., Saito, Y., Koda, Y., Sato, K. & Sakakibara, Y. Convolutional neural network based on smiles representation of compounds for detecting chemical motif. BMC Bioinformatics 19 , 83–94 (2018).

Gómez-Bombarelli, R. et al. Automatic chemical design using a data-driven continuous representation of molecules. ACS Central Sci. 4 , 268–276 (2018).

Liu, R. et al. Deep learning for chemical compound stability prediction . In Proceedings of ACM SIGKDD workshop on large-scale deep learning for data mining (DL-KDD) , 1–7. https://rosanneliu.com/publication/kdd/ (ACM SIGKDD, 2016).

Jha, D. et al. Elemnet: Deep learning the chem. mater. from only elemental composition. Sci. Rep. 8 , 1–13 (2018).

Agrawal, A. et al. Exploration of data science techniques to predict fatigue strength of steel from composition and processing parameters. Integr. Mater. Manuf. Innov. 3 , 90–108 (2014).

Agrawal, A. & Choudhary, A. A fatigue strength predictor for steels using ensemble data mining: steel fatigue strength predictor . In Proceedings of the 25th ACM International on Conference on information and knowledge management , 2497–2500. https://doi.org/10.1145/2983323.2983343 (2016).

Agrawal, A. & Choudhary, A. An online tool for predicting fatigue strength of steel alloys based on ensemble data mining. Int. J. Fatigue 113 , 389–400 (2018).

Agrawal, A., Saboo, A., Xiong, W., Olson, G. & Choudhary, A. Martensite start temperature predictor for steels using ensemble data mining . in 2019 IEEE International Conference on Data Science and Advanced Analytics (DSAA) , 521–530 (IEEE, 2019).

Meredig, B. et al. Combinatorial screening for new materials in unconstrained composition space with machine learning. Phys. Rev. B 89 , 094104 (2014).

Agrawal, A., Meredig, B., Wolverton, C. & Choudhary, A. A formation energy predictor for crystalline materials using ensemble data mining . in 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW) , 1276–1279 (IEEE, 2016).

Furmanchuk, A., Agrawal, A. & Choudhary, A. Predictive analytics for crystalline materials: bulk modulus. RSC Adv. 6 , 95246–95251 (2016).

Furmanchuk, A. et al. Prediction of seebeck coefficient for compounds without restriction to fixed stoichiometry: A machine learning approach. J. Comput. Chem. 39 , 191–202 (2018).

Ward, L., Agrawal, A., Choudhary, A. & Wolverton, C. A general-purpose machine learning framework for predicting properties of inorganic materials. npj Comput. Mater. 2 , 1–7 (2016).

Ward, L. et al. Matminer: An open source toolkit for materials data mining. Comput. Mater. Sci. 152 , 60–69 (2018).

Jha, D. et al. Irnet: A general purpose deep residual regression framework for materials discovery . In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining , 2385–2393. https://arxiv.org/abs/1907.03222 (2019).

Jha, D. et al. Enabling deeper learning on big data for materials informatics applications. Sci. Rep. 11 , 1–12 (2021).

Goodall, R. E. & Lee, A. A. Predicting materials properties without crystal structure: Deep representation learning from stoichiometry. Nat. Commun. 11 , 1–9 (2020).

NIMS. Superconducting material database (supercon) . https://supercon.nims.go.jp/ (2021).

Stanev, V. et al. Machine learning modeling of superconducting critical temperature. npj Comput. Mater. 4 , 1–14 (2018).

Gupta, V. et al. Cross-property deep transfer learning framework for enhanced predictive analytics on small materials data. Nat. Commun . 12 , 1–10 (2021).

Himanen, L. et al. Dscribe: Library of descriptors for machine learning in materials science. Computer Phys. Commun. 247 , 106949 (2020).

Bartel, C. J. et al. A critical examination of compound stability predictions from machine-learned formation energies. npj Comput. Mater. 6 , 1–11 (2020).

Choudhary, K. et al. High-throughput density functional perturbation theory and machine learning predictions of infrared, piezoelectric, and dielectric responses. npj Comput. Mater. 6 , 1–13 (2020).

Zheng, C. et al. Automated generation and ensemble-learned matching of X-ray absorption spectra. npj Comput. Mater. 4 , 1–9 (2018).

Mathew, K. et al. High-throughput computational x-ray absorption spectroscopy. Sci. Data 5 , 1–8 (2018).

Chen, Y. et al. Database of ab initio l-edge x-ray absorption near edge structure. Sci. Data 8 , 1–8 (2021).

Lafuente, B., Downs, R. T., Yang, H. & Stone, N. In Highlights in mineralogical crystallography 1–30 (De Gruyter (O), 2015).

El Mendili, Y. et al. Raman open database: first interconnected raman–x-ray diffraction open-access resource for material identification. J. Appl. Crystallogr. 52 , 618–625 (2019).

Fremout, W. & Saverwyns, S. Identification of synthetic organic pigments: the role of a comprehensive digital raman spectral library. J. Raman Spectrosc. 43 , 1536–1544 (2012).

Huck, P. & Persson, K. A. Mpcontribs: user contributed data to the materials project database . https://docs.mpcontribs.org/ (2019).

Yang, L. et al. A cloud platform for atomic pair distribution function analysis: Pdfitc. Acta Crystallogr. A 77 , 2–6 (2021).

Park, W. B. et al. Classification of crystal structure using a convolutional neural network. IUCrJ 4 , 486–494 (2017).

Hellenbrandt, M. The Inorganic Crystal Structure Database (ICSD)—present and future. Crystallogr. Rev. 10 , 17–22 (2004).

Zaloga, A. N., Stanovov, V. V., Bezrukova, O. E., Dubinin, P. S. & Yakimov, I. S. Crystal symmetry classification from powder X-ray diffraction patterns using a convolutional neural network. Mater. Today Commun. 25 , 101662 (2020).

Lee, J.-W., Park, W. B., Lee, J. H., Singh, S. P. & Sohn, K.-S. A deep-learning technique for phase identification in multiphase inorganic compounds using synthetic XRD powder patterns. Nat. Commun. 11 , 86 (2020).

Wang, H. et al. Rapid identification of X-ray diffraction patterns based on very limited data by interpretable convolutional neural networks. J. Chem. Inf. Model. 60 , 2004–2011 (2020).

Dong, H. et al. A deep convolutional neural network for real-time full profile analysis of big powder diffraction data. npj Comput. Mater. 7 , 1–9 (2021).

Aguiar, J. A., Gong, M. L. & Tasdizen, T. Crystallographic prediction from diffraction and chemistry data for higher throughput classification using machine learning. Comput. Mater. Sci. 173 , 109409 (2020).

Maffettone, P. M. et al. Crystallography companion agent for high-throughput materials discovery. Nat. Comput. Sci. 1 , 290–297 (2021).

Oviedo, F. et al. Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks. npj Comput. Mater. 5 , 1–9 (2019).

Liu, C.-H. et al. Validation of non-negative matrix factorization for rapid assessment of large sets of atomic pair-distribution function (pdf) data. J. Appl. Crystallogr. 54 , 768–775 (2021).

Rakita, Y. et al. Studying heterogeneities in local nanostructure with scanning nanostructure electron microscopy (snem). arXiv https://arxiv.org/abs/2110.03589 (2021).

Timoshenko, J., Lu, D., Lin, Y. & Frenkel, A. I. Supervised machine-learning-based determination of three-dimensional structure of metallic nanoparticles. J. Phys. Chem Lett. 8 , 5091–5098 (2017).

Timoshenko, J. et al. Subnanometer substructures in nanoassemblies formed from clusters under a reactive atmosphere revealed using machine learning. J. Phys. Chem C 122 , 21686–21693 (2018).

Timoshenko, J. et al. Neural network approach for characterizing structural transformations by X-ray absorption fine structure spectroscopy. Phys. Rev. Lett. 120 , 225502 (2018).

Zheng, C., Chen, C., Chen, Y. & Ong, S. P. Random forest models for accurate identification of coordination environments from X-ray absorption near-edge structure. Patterns 1 , 100013 (2020).

Torrisi, S. B. et al. Random forest machine learning models for interpretable X-ray absorption near-edge structure spectrum-property relationships. npj Comput. Mater. 6 , 1–11 (2020).

Andrejevic, N., Andrejevic, J., Rycroft, C. H. & Li, M. Machine learning spectral indicators of topology. arXiv preprint at https://arxiv.org/abs/2003.00994 (2020).

Madden, M. G. & Ryder, A. G. Machine learning methods for quantitative analysis of raman spectroscopy data . in Opto-Ireland 2002: Optics and Photonics Technologies and Applications , Vol. 4876, 1130–1139 (International Society for Optics and Photonics, 2003).

Conroy, J., Ryder, A. G., Leger, M. N., Hennessey, K. & Madden, M. G. Qualitative and quantitative analysis of chlorinated solvents using Raman spectroscopy and machine learning . in Opto-Ireland 2005: Optical Sensing and Spectroscopy, Vol. 5826, 131–142 (International Society for Optics and Photonics, 2005).

Acquarelli, J. et al. Convolutional neural networks for vibrational spectroscopic data analysis. Anal. Chim. Acta 954 , 22–31 (2017).

O’Connell, M.-L., Howley, T., Ryder, A. G., Leger, M. N. & Madden, M. G. Classification of a target analyte in solid mixtures using principal component analysis, support vector machines, and Raman spectroscopy . in Opto-Ireland 2005: Optical Sensing and Spectroscopy , Vol. 5826, 340–350 (International Society for Optics and Photonics, 2005).

Zhao, J., Chen, Q., Huang, X. & Fang, C. H. Qualitative identification of tea categories by near infrared spectroscopy and support vector machine. J. Pharm. Biomed. Anal. 41 , 1198–1204 (2006).

Liu, J. et al. Deep convolutional neural networks for Raman spectrum recognition: a unified solution. Analyst 142 , 4067–4074 (2017).

Yang, J. et al. Deep learning for vibrational spectral analysis: Recent progress and a practical guide. Anal. Chim. Acta 1081 , 6–17 (2019).

Selzer, P., Gasteiger, J., Thomas, H. & Salzer, R. Rapid access to infrared reference spectra of arbitrary organic compounds: scope and limitations of an approach to the simulation of infrared spectra by neural networks. Chem. Euro. J. 6 , 920–927 (2000).

Ghosh, K. et al. Deep learning spectroscopy: neural networks for molecular excitation spectra. Adv. Sci. 6 , 1801367 (2019).

Kostka, T., Selzer, P. & Gasteiger, J. A combined application of reaction prediction and infrared spectra simulation for the identification of degradation products of s-triazine herbicides. Chemistry 7 , 2254–2260 (2001).

Mahmoud, C. B., Anelli, A., Csányi, G. & Ceriotti, M. Learning the electronic density of states in condensed matter. Phys. Rev. B 102 , 235130 (2020).

Chen, Z. et al. Direct prediction of phonon density of states with Euclidean neural networks. Adv. Sci. 8 , 2004214 (2021).

Kong, S. et al. Density of states prediction for materials discovery via contrastive learning from probabilistic embeddings. arXiv . https://arxiv.org/abs/2110.11444 (2021).

Carbone, M. R., Topsakal, M., Lu, D. & Yoo, S. Machine-learning X-ray absorption spectra to quantitative accuracy. Phys. Rev. Lett. 124 , 156401 (2020).

Rehr, J. J., Kas, J. J., Vila, F. D., Prange, M. P. & Jorissen, K. Parameter-free calculations of X-ray spectra with FEFF9. Phys. Chem. Chem. Phys. 12 , 5503–5513 (2010).

Rankine, C. D., Madkhali, M. M. M. & Penfold, T. J. A deep neural network for the rapid prediction of X-ray absorption spectra. J. Phys. Chem A 124 , 4263–4270 (2020).

Fung, V., Hu, G., Ganesh, P. & Sumpter, B. G. Machine learned features from density of states for accurate adsorption energy prediction. Nat. Commun. 12 , 88 (2021).

Hammer, B. & Nørskov, J. Theoretical surface science and catalysis-calculations and concepts. Adv. Catal. Impact Surface Sci. Catal. 45 , 71–129 (2000).

Kaundinya, P. R., Choudhary, K. & Kalidindi, S. R. Prediction of the electron density of states for crystalline compounds with atomistic line graph neural networks (alignn). arXiv. https://arxiv.org/abs/2201.08348 (2022).

Stein, H. S., Soedarmadji, E., Newhouse, P. F., Guevarra, D. & Gregoire, J. M. Synthesis, optical imaging, and absorption spectroscopy data for 179072 metal oxides. Sci. Data 6 , 9 (2019).

Choudhary, A. et al. Graph neural network predictions of metal organic framework co2 adsorption properties. arXiv . https://arxiv.org/abs/2112.10231 (2021).

Anderson, R., Biong, A. & Gómez-Gualdrón, D. A. Adsorption isotherm predictions for multiple molecules in mofs using the same deep learning model. J. Chem. Theory Comput. 16 , 1271–1283 (2020).

Krizhevsky, A., Sutskever, I. & Hinton, G. E. Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25 , 1097–1105 (2012).

Varela, M. et al. Materials characterization in the aberration-corrected scanning transmission electron microscope. Annu. Rev. Mater. Res. 35 , 539–569 (2005).

Holm, E. A. et al. Overview: Computer vision and machine learning for microstructural characterization and analysis. Metal. Mater Trans. A 51 , 5985–5999 (2020).

Modarres, M. H. et al. Neural network for nanoscience scanning electron microscope image recognition. Sci. Rep. 7 , 1–12 (2017).

Gopalakrishnan, K., Khaitan, S. K., Choudhary, A. & Agrawal, A. Deep convolutional neural networks with transfer learning for computer vision-based data-driven pavement distress detection. Construct. Build. Mater. 157 , 322–330 (2017).

Gopalakrishnan, K., Gholami, H., Vidyadharan, A., Choudhary, A. & Agrawal, A. Crack damage detection in unmanned aerial vehicle images of civil infrastructure using pre-trained deep learning model. Int. J. Traffic Transp. Eng . 8 , 1–14 (2018).

Yang, Z. et al. Data-driven insights from predictive analytics on heterogeneous experimental data of industrial magnetic materials . In IEEE International Conference on Data Mining Workshops (ICDMW) , 806–813. https://doi.org/10.1109/ICDMW.2019.00119 (IEEE Computer Society, 2019).

Yang, Z. et al. Heterogeneous feature fusion based machine learning on shallow-wide and heterogeneous-sparse industrial datasets . In 25th International Conference on Pattern Recognition Workshops, ICPR 2020 , 566–577. https://doi.org/10.1007/978-3-030-68799-1_41 (Springer Science and Business Media Deutschland GmbH, 2021).

Ziletti, A., Kumar, D., Scheffler, M. & Ghiringhelli, L. M. Insightful classification of crystal structures using deep learning. Nat. Commun. 9 , 2775 (2018).

Choudhary, K. et al. Computational scanning tunneling microscope image database. Sci. Data 8 , 1–9 (2021).

Liu, R., Agrawal, A., Liao, W.-k., Choudhary, A. & De Graef, M. Materials discovery: Understanding polycrystals from large-scale electron patterns . in 2016 IEEE International Conference on Big Data (Big Data) , 2261–2269 (IEEE, 2016).

Jha, D. et al. Extracting grain orientations from EBSD patterns of polycrystalline materials using convolutional neural networks. Microsc. Microanal. 24 , 497–502 (2018).

Kaufmann, K., Zhu, C., Rosengarten, A. S. & Vecchio, K. S. Deep neural network enabled space group identification in EBSD. Microsc. Microanal. 26 , 447–457 (2020).

Yang, Z. et al. Deep learning based domain knowledge integration for small datasets: Illustrative applications in materials informatics . in 2019 International Joint Conference on Neural Networks (IJCNN) , 1–8 (IEEE, 2019).

Yang, Z. et al. Learning to predict crystal plasticity at the nanoscale: Deep residual networks and size effects in uniaxial compression discrete dislocation simulations. Sci. Rep. 10 , 1–14 (2020).

Decost, B. L. et al. Uhcsdb: Ultrahigh carbon steel micrograph database. Integr. Mater. Manuf. Innov. 6 , 197–205 (2017).

Decost, B. L., Lei, B., Francis, T. & Holm, E. A. High throughput quantitative metallography for complex microstructures using deep learning: a case study in ultrahigh carbon steel. Microsc. Microanal. 25 , 21–29 (2019).

Stan, T., Thompson, Z. T. & Voorhees, P. W. Optimizing convolutional neural networks to perform semantic segmentation on large materials imaging datasets: X-ray tomography and serial sectioning. Materials Characterization 160 , 110119 (2020).

Madsen, J. et al. A deep learning approach to identify local structures in atomic-resolution transmission electron microscopy images. Adv. Theory Simulations 1 , 1800037 (2018).

Maksov, A. et al. Deep learning analysis of defect and phase evolution during electron beam-induced transformations in ws 2. npj Comput. Mater. 5 , 1–8 (2019).

Yang, S.-H. et al. Deep learning-assisted quantification of atomic dopants and defects in 2d materials. Adv. Sci. https://doi.org/10.1002/advs.202101099 (2021).

Roberts, G. et al. Deep learning for semantic segmentation of defects in advanced stem images of steels. Sci. Rep. 9 , 1–12 (2019).

Kusche, C. et al. Large-area, high-resolution characterisation and classification of damage mechanisms in dual-phase steel using deep learning. PLoS ONE 14 , e0216493 (2019).

Vlcek, L. et al. Learning from imperfections: predicting structure and thermodynamics from atomic imaging of fluctuations. ACS Nano 13 , 718–727 (2019).

Ziatdinov, M., Maksov, A. & Kalinin, S. V. Learning surface molecular structures via machine vision. npj Comput. Mater. 3 , 1–9 (2017).

Ovchinnikov, O. S. et al. Detection of defects in atomic-resolution images of materials using cycle analysis. Adv. Struct. Chem. Imaging 6 , 3 (2020).

Li, W., Field, K. G. & Morgan, D. Automated defect analysis in electron microscopic images. npj Comput. Mater. 4 , 1–9 (2018).

Cohn, R. et al. Instance segmentation for direct measurements of satellites in metal powders and automated microstructural characterization from image data. JOM 73 , 2159–2172 (2021).

de Haan, K., Ballard, Z. S., Rivenson, Y., Wu, Y. & Ozcan, A. Resolution enhancement in scanning electron microscopy using deep learning. Sci. Rep. 9 , 1–7 (2019).

Ede, J. M. & Beanland, R. Partial scanning transmission electron microscopy with deep learning. Sci. Rep. 10 , 1–10 (2020).

Rashidi, M. & Wolkow, R. A. Autonomous scanning probe microscopy in situ tip conditioning through machine learning. ACS Nano 12 , 5185–5189 (2018).

Scime, L., Siddel, D., Baird, S. & Paquit, V. Layer-wise anomaly detection and classification for powder bed additive manufacturing processes: A machine-agnostic algorithm for real-time pixel-wise semantic segmentation. Addit. Manufact. 36 , 101453 (2020).

Eppel, S., Xu, H., Bismuth, M. & Aspuru-Guzik, A. Computer vision for recognition of materials and vessels in chemistry lab settings and the Vector-LabPics Data Set. ACS Central Sci. 6 , 1743–1752 (2020).

Yang, Z. et al. Deep learning approaches for mining structure-property linkages in high contrast composites from simulation datasets. Comput. Mater. Sci. 151 , 278–287 (2018).

Cecen, A., Dai, H., Yabansu, Y. C., Kalidindi, S. R. & Song, L. Material structure-property linkages using three-dimensional convolutional neural networks. Acta Mater. 146 , 76–84 (2018).

Yang, Z. et al. Establishing structure-property localization linkages for elastic deformation of three-dimensional high contrast composites using deep learning approaches. Acta Mater. 166 , 335–345 (2019).

Goetz, A. et al. Addressing materials’ microstructure diversity using transfer learning. arXiv . arXiv-2107. https://arxiv.org/abs/2107.13841 (2021).

Kitahara, A. R. & Holm, E. A. Microstructure cluster analysis with transfer learning and unsupervised learning. Integr. Mater. Manuf. Innov. 7 , 148–156 (2018).

Larmuseau, M. et al. Compact representations of microstructure images using triplet networks. npj Comput. Mater. 2020 6:1 6 , 1–11 (2020).

Li, X. et al. A deep adversarial learning methodology for designing microstructural material systems . in International Design Engineering Technical Conferences and Computers and Information in Engineering Conference , Vol. 51760, V02BT03A008 (American Society of Mechanical Engineers, 2018).

Yang, Z. et al. Microstructural materials design via deep adversarial learning methodology. J. Mech. Des. 140 , 111416 (2018).

Yang, Z. et al. A general framework combining generative adversarial networks and mixture density networks for inverse modeling in microstructural materials design. arXiv . https://arxiv.org/abs/2101.10553 (2021).

Hsu, T. et al. Microstructure generation via generative adversarial network for heterogeneous, topologically complex 3d materials. JOM 73 , 90–102 (2020).

Chun, S. et al. Deep learning for synthetic microstructure generation in a materials-by-design framework for heterogeneous energetic materials. Sci. Rep. 10 , 1–15 (2020).

Dai, M., Demirel, M. F., Liang, Y. & Hu, J.-M. Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials. npj Comput. Mater. 7 , 1–9 (2021).

Cohn, R. & Holm, E. Neural message passing for predicting abnormal grain growth in Monte Carlo simulations of microstructural evolution. arXiv. https://arxiv.org/abs/2110.09326v1 (2021).

Plimpton, S. et al. SPPARKS Kinetic Monte Carlo Simulator . https://spparks.github.io/index.html . (2021).

Plimpton, S. et al. Crossing the mesoscale no-man’s land via parallel kinetic Monte Carlo. Tech. Rep . https://doi.org/10.2172/966942 (2009).

Xue, N. Steven bird, evan klein and edward loper. natural language processing with python. oreilly media, inc.2009. isbn: 978-0-596-51649-9. Nat. Lang. Eng. 17 , 419–424 (2010).

Honnibal, M. & Montani, I. spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing. https://doi.org/10.5281/zenodo.3358113 (2017).

Gardner, M. et al. Allennlp: A deep semantic natural language processing platform. arXiv. https://arxiv.org/abs/1803.07640 (2018).

Tshitoyan, V. et al. Unsupervised word embeddings capture latent knowledge from materials science literature. Nature 571 , 95–98 (2019).

Kononova, O. et al. Opportunities and challenges of text mining in aterials research. iScience 24 , 102155 (2021).

Olivetti, E. A. et al. Data-driven materials research enabled by natural language processing and information extraction. Appl. Phys. Rev. 7 , 041317 (2020).

Swain, M. C. & Cole, J. M. Chemdataextractor: a toolkit for automated extraction of chemical information from the scientific literature. J. Chem. Inf. Model. 56 , 1894–1904 (2016).

Park, S. et al. Text mining metal–organic framework papers. J. Chem. Inf. Model. 58 , 244–251 (2018).

Court, C. J. & Cole, J. M. Auto-generated materials database of curie and néel temperatures via semi-supervised relationship extraction. Sci. Data 5 , 1–12 (2018).

Huang, S. & Cole, J. M. A database of battery materials auto-generated using chemdataextractor. Sci. Data 7 , 1–13 (2020).

Beard, E. J., Sivaraman, G., Vázquez-Mayagoitia, Á., Vishwanath, V. & Cole, J. M. Comparative dataset of experimental and computational attributes of uv/vis absorption spectra. Sci. Data 6 , 1–11 (2019).

Tayfuroglu, O., Kocak, A. & Zorlu, Y. In silico investigation into h2 uptake in mofs: combined text/data mining and structural calculations. Langmuir 36 , 119–129 (2019).

Weston, L. et al. Named entity recognition and normalization applied to large-scale information extraction from the materials science literature. J. Chem. Inf. Model. 59 , 3692–3702 (2019).

Vaucher, A. C. et al. Automated extraction of chemical synthesis actions from experimental procedures. Nat. Commun. 11 , 1–11 (2020).

He, T. et al. Similarity of precursors in solid-state synthesis as text-mined from scientific literature. Chem. Mater. 32 , 7861–7873 (2020).

Kononova, O. et al. Text-mined dataset of inorganic materials synthesis recipes. Sci. Data 6 , 1–11 (2019).

Kim, E. et al. Materials synthesis insights from scientific literature via text extraction and machine learning. Chem. Mater. 29 , 9436–9444 (2017).

Kim, E., Huang, K., Jegelka, S. & Olivetti, E. Virtual screening of inorganic materials synthesis parameters with deep learning. npj Comput. Mater. 3 , 1–9 (2017).

Kim, E. et al. Inorganic materials synthesis planning with literature-trained neural networks. J. Chem. Inf. Model. 60 , 1194–1201 (2020).

de Castro, P. B. et al. Machine-learning-guided discovery of the gigantic magnetocaloric effect in hob 2 near the hydrogen liquefaction temperature. NPG Asia Mater. 12 , 1–7 (2020).

Cooper, C. B. et al. Design-to-device approach affords panchromatic co-sensitized solar cells. Adv. Energy Mater. 9 , 1802820 (2019).

Yang, X., Dai, Z., Zhao, Y., Liu, J. & Meng, S. Low lattice thermal conductivity and excellent thermoelectric behavior in li3sb and li3bi. J. Phys. Condens. Matter 30 , 425401 (2018).

Wang, Y., Gao, Z. & Zhou, J. Ultralow lattice thermal conductivity and electronic properties of monolayer 1t phase semimetal site2 and snte2. Phys. E 108 , 53–59 (2019).

Jong, U.-G., Yu, C.-J., Kye, Y.-H., Hong, S.-N. & Kim, H.-G. Manifestation of the thermoelectric properties in ge-based halide perovskites. Phys. Rev. Mater. 4 , 075403 (2020).

Yamamoto, K., Narita, G., Yamasaki, J. & Iikubo, S. First-principles study of thermoelectric properties of mixed iodide perovskite cs (b, b’) i3 (b, b’= ge, sn, and pb). J. Phys. Chem. Solids 140 , 109372 (2020).

Viennois, R. et al. Anisotropic low-energy vibrational modes as an effect of cage geometry in the binary barium silicon clathrate b a 24 s i 100. Phys. Rev. B 101 , 224302 (2020).

Haque, E. Effect of electron-phonon scattering, pressure and alloying on the thermoelectric performance of tmcu _3 ch _4(tm= v, nb, ta; ch= s, se, te). arXiv . https://arxiv.org/abs/2010.08461 (2020).

Yahyaoglu, M. et al. Phase-transition-enhanced thermoelectric transport in rickardite mineral cu3–x te2. Chem. Mater. 33 , 1832–1841 (2021).

Ho, D., Shkolnik, A. S., Ferraro, N. J., Rizkin, B. A. & Hartman, R. L. Using word embeddings in abstracts to accelerate metallocene catalysis polymerization research. Computers Chem. Eng. 141 , 107026 (2020).

Abdar, M. et al. A review of uncertainty quantification in deep learning: techniques, applications and challenges. Inf. Fusion . 76 , 243–297 (2021).

Mi, Lu, et al. Training-free uncertainty estimation for dense regression: Sensitivityas a surrogate. arXiv . preprint at arXiv:1910.04858. https://arxiv.org/abs/1910.04858 (2019).

Teye, M., Azizpour, H. & Smith, K. Bayesian uncertainty estimation for batch normalized deep networks . in International Conference on Machine Learning , 4907–4916 (PMLR, 2018).

Zhang, J., Kailkhura, B. & Han, T. Y.-J. Leveraging uncertainty from deep learning for trustworthy material discovery workflows. ACS Omega 6 , 12711–12721 (2021).

Meredig, B. et al. Can machine learning identify the next high-temperature superconductor? examining extrapolation performance for materials discovery. Mol. Syst. Des. Eng. 3 , 819–825 (2018).

Zhang, J., Kailkhura, B. & Han, T. Y.-J. Mix-n-match: Ensemble and compositional methods for uncertainty calibration in deep learning . in International Conference on Machine Learning , 11117–11128 (PMLR, 2020).

Seoh, R. Qualitative analysis of monte carlo dropout. arXiv. https://arxiv.org/abs/2007.01720 (2020).

Gal, Y. & Ghahramani, Z. Dropout as a bayesian approximation: Representing model uncertainty in deep learning . in international conference on machine learning , 1050–1059 (PMLR, 2016).

Jain, S., Liu, G., Mueller, J. & Gifford, D. Maximizing overall diversity for improved uncertainty estimates in deep ensembles . In Proceedings of the AAAI Conference on Artificial Intelligence , 34 , 4264–4271. https://doi.org/10.1609/aaai.v34i04.5849 (2020).

Ganaie, M. et al. Ensemble deep learning: a review. arXiv . https://arxiv.org/abs/2104.02395 (AAAI Technical Track: Machine Learning, 2021).

Fort, S., Hu, H. & Lakshminarayanan, B. Deep ensembles: a loss landscape perspective. arXiv. https://arxiv.org/abs/1912.02757 (2019).

Lakshminarayanan, B., Pritzel, A. & Blundell, C. Simple and scalable predictive uncertainty estimation using deep ensembles. arXiv. https://arxiv.org/abs/1612.01474 (2016).

Moon, S. J., Jeon, J.-J., Lee, J. S. H. & Kim, Y. Learning multiple quantiles with neural networks. J. Comput. Graph. Stat. 30 , 1–11. https://doi.org/10.1080/10618600.2021.1909601 (2021).

Rasmussen, C. E. Summer School on Machine Learning , 63–71 (Springer, 2003).

Hegde, P., Heinonen, M., Lähdesmäki, H. & Kaski, S. Deep learning with differential gaussian process flows. arXiv. https://arxiv.org/abs/1810.04066 (2018).

Wilson, A. G., Hu, Z., Salakhutdinov, R. & Xing, E. P. Deep kernel learning. in Artificial intelligence and statistics , 370–378 (PMLR, 2016).

Hegde, V. I. et al. Reproducibility in high-throughput density functional theory: a comparison of aflow, materials project, and oqmd. arXiv. https://arxiv.org/abs/2007.01988 (2020).

Ying, R., Bourgeois, D., You, J., Zitnik, M. & Leskovec, J. Gnnexplainer: Generating explanations for graph neural networks. Adv. Neural Inf. Process. Syst. 32 , 9240 (2019).

Roch, L. M. et al. Chemos: orchestrating autonomous experimentation. Sci. Robot. 3 , eaat5559 (2018).

Szymanski, N. et al. Toward autonomous design and synthesis of novel inorganic materials. Mater. Horiz. 8 , 2169–2198. https://doi.org/10.1039/D1MH00495F (2021).

MacLeod, B. P. et al. Self-driving laboratory for accelerated discovery of thin-film materials. Sci. Adv. 6 , eaaz8867 (2020).

Stach, E. A. et al. Autonomous experimentation systems for materials development: a community perspective. Matter https://www.cell.com/matter/fulltext/S2590-2385(21)00306-4 (2021).

Rakita, Y. et al. Active reaction control of cu redox state based on real-time feedback from i n situ synchrotron measurements. J. Am. Chem. Soc. 142 , 18758–18762 (2020).

Chmiela, S. et al. Machine learning of accurate energy-conserving molecular force fields. Sci. Adv. 3 , e1603015 (2017).

Thomas, R. S. et al. The us federal tox21 program: a strategic and operational plan for continued leadership. Altex 35 , 163 (2018).

Russell Johnson, N. Nist computational chemistry comparison and benchmark database . In The 4th Joint Meeting of the US Sections of the Combustion Institute . https://ci.confex.com/ci/2005/techprogram/P1309.HTM (2005).

Lopez, S. A. et al. The harvard organic photovoltaic dataset. Sci. Data 3 , 1–7 (2016).

Johnson, R. D. et al. Nist computational chemistry comparison and benchmark database . http://srdata.nist.gov/cccbdb (2006).

Mobley, D. L. & Guthrie, J. P. Freesolv: a database of experimental and calculated hydration free energies, with input files. J. Computer Aided Mol. Des. 28 , 711–720 (2014).

Andersen, C. W. et al. Optimade: an api for exchanging materials data. arXiv. https://arxiv.org/abs/2103.02068 (2021).

Chanussot, L. et al. Open catalyst 2020 (oc20) dataset and community challenges. ACS Catal. 11 , 6059–6072 (2021).

Dunn, A., Wang, Q., Ganose, A., Dopp, D. & Jain, A. Benchmarking materials property prediction methods: the matbench test set and automatminer reference algorithm. npj Comput. Mater. 6 , 1–10 (2020).

Talirz, L. et al. Materials cloud, a platform for open computational science. Sci. Data 7 , 1–12 (2020).

Chung, Y. G. et al. Advances, updates, and analytics for the computation-ready, experimental metal–organic framework database: Core mof 2019. J. Chem. Eng. Data 64 , 5985–5998 (2019).

Sussman, J. L. et al. Protein data bank (pdb): database of three-dimensional structural information of biological macromolecules. Acta Crystallogr. Sec. D Biol. Crystallogr. 54 , 1078–1084 (1998).

Benson, M. L. et al. Binding moad, a high-quality protein–ligand database. Nucleic Acids Res. 36 , D674–D678 (2007).

Fung, V., Zhang, J., Juarez, E. & Sumpter, B. G. Benchmarking graph neural networks for materials chemistry. npj Comput. Mater. 7 , 1–8 (2021).

Louis, S.-Y. et al. Graph convolutional neural networks with global attention for improved materials property prediction. Phys. Chem. Chem. Phys. 22 , 18141–18148 (2020).

Khorshidi, A. & Peterson, A. A. Amp: A modular approach to machine learning in atomistic simulations. Computer Phys. Commun. 207 , 310–324 (2016).

Yao, K., Herr, J. E., Toth, D. W., Mckintyre, R. & Parkhill, J. The tensormol-0.1 model chemistry: a neural network augmented with long-range physics. Chem. Sci. 9 , 2261–2269 (2018).

Doerr, S. et al. Torchmd: A deep learning framework for molecular simulations. J. Chem. Theory Comput. 17 , 2355–2363 (2021).

Kolb, B., Lentz, L. C. & Kolpak, A. M. Discovering charge density functionals and structure-property relationships with prophet: A general framework for coupling machine learning and first-principles methods. Sci. Rep. 7 , 1–9 (2017).

Zhang, L., Han, J., Wang, H., Car, R. & Weinan, E. Deep potential molecular dynamics: a scalable model with the accuracy of quantum mechanics. Phys. Rev. Lett. 120 , 143001 (2018).

Geiger, M. et al. e3nn/e3nn: 2021-06-21 . https://doi.org/10.5281/zenodo.5006322 (2021).

Duvenaud, D. K. et al. Convolutional networks on graphs for learning molecular fingerprints (eds. Cortes, C., Lawrence, N. D., Lee, D. D., Sugiyama, M. & Garnett, R.) in Adv. Neural Inf. Process. Syst. 28 2224–2232 (Curran Associates, Inc., 2015).

Li, X. et al. Deepchemstable: Chemical stability prediction with an attention-based graph convolution network. J. Chem. Inf. Model. 59 , 1044–1049 (2019).

Wu, Z. et al. MoleculeNet: A benchmark for molecular machine learning. Chem. Sci. 9 , 513–530 (2018).

Wang, A. Y.-T., Kauwe, S. K., Murdock, R. J. & Sparks, T. D. Compositionally restricted attention-based network for materials property predictions. npj Comput. Mater. 7 , 77 (2021).

Zhou, Q. et al. Learning atoms for materials discovery. Proc. Natl Acad. Sci. USA 115 , E6411–E6417 (2018).

O’Boyle, N. & Dalke, A. Deepsmiles: An adaptation of smiles for use in machine-learning of chemical structures. ChemRxiv https://doi.org/10.26434/chemrxiv.7097960.v1 (2018).

Green, H., Koes, D. R. & Durrant, J. D. Deepfrag: a deep convolutional neural network for fragment-based lead optimization. Chem. Sci. 12 , 8036–8047. https://doi.org/10.1039/D1SC00163A (2021).

Elhefnawy, W., Li, M., Wang, J. & Li, Y. Deepfrag-k: a fragment-based deep learning approach for protein fold recognition. BMC Bioinformatics 21 , 203 (2020).

Paul, A. et al. Chemixnet: Mixed dnn architectures for predicting chemical properties using multiple molecular representations. arXiv . https://arxiv.org/abs/1811.08283 (2018).

Paul, A. et al. Transfer learning using ensemble neural networks for organic solar cell screening . in 2019 International Joint Conference on Neural Networks (IJCNN) , 1–8 (IEEE, 2019).

Choudhary, K. et al. Computational screening of high-performance optoelectronic materials using optb88vdw and tb-mbj formalisms. Sci. Data 5 , 1–12 (2018).

Wong-Ng, W., McMurdie, H., Hubbard, C. & Mighell, A. D. Jcpds-icdd research associateship (cooperative program with nbs/nist). J. Res. Natl Inst. Standards Technol. 106 , 1013 (2001).

Belsky, A., Hellenbrandt, M., Karen, V. L. & Luksch, P. New developments in the inorganic crystal structure database (icsd): accessibility in support of materials research and design. Acta Crystallogr. Sec. B Struct. Sci. 58 , 364–369 (2002).

Gražulis, S. et al. Crystallography Open Database—an open-access collection of crystal structures. J. Appl. Crystallogr. 42 , 726–729 (2009).

Linstrom, P. J. & Mallard, W. G. The nist chemistry webbook: a chemical data resource on the internet. J. Chem. Eng. Data 46 , 1059–1063 (2001).

Saito, T. et al. Spectral database for organic compounds (sdbs). (National Institute of Advanced Industrial Science and Technology (AIST), 2006).

Steinbeck, C., Krause, S. & Kuhn, S. Nmrshiftdb constructing a free chemical information system with open-source components. J. Chem. inf. Computer Sci. 43 , 1733–1739 (2003).

Fung, V., Hu, G., Ganesh, P. & Sumpter, B. G. Machine learned features from density of states for accurate adsorption energy prediction. Nat. Commun. 12 , 1–11 (2021).

Kong, S., Guevarra, D., Gomes, C. P. & Gregoire, J. M. Materials representation and transfer learning for multi-property prediction. arXiv . https://arxiv.org/abs/2106.02225 (2021).

Bang, K., Yeo, B. C., Kim, D., Han, S. S. & Lee, H. M. Accelerated mapping of electronic density of states patterns of metallic nanoparticles via machine-learning. Sci. Rep . 11 , 1–11 (2021).

Chen, D. et al. Automating crystal-structure phase mapping by combining deep learning with constraint reasoning. Nat. Machine Intell. 3 , 812–822 (2021).

Ophus, C. A fast image simulation algorithm for scanning transmission electron microscopy. Adv. Struct. Chem. imaging 3 , 1–11 (2017).

Aversa, R., Modarres, M. H., Cozzini, S., Ciancio, R. & Chiusole, A. The first annotated set of scanning electron microscopy images for nanoscience. Sci. Data 5 , 1–10 (2018).

Ziatdinov, M. et al. Causal analysis of competing atomistic mechanisms in ferroelectric materials from high-resolution scanning transmission electron microscopy data. npj Comput. Mater. 6 , 1–9 (2020).

Souza, A. L. F. et al. Deepfreak: Learning crystallography diffraction patterns with automated machine learning. arXiv. http://arxiv.org/abs/1904.11834 (2019).

Scime, L. et al. Layer-wise imaging dataset from powder bed additive manufacturing processes for machine learning applications (peregrine v2021-03). Tech. Rep . https://www.osti.gov/biblio/1779073 (2021).

Somnath, S., Smith, C. R., Laanait, N., Vasudevan, R. K. & Jesse, S. Usid and pycroscopy–open source frameworks for storing and analyzing imaging and spectroscopy data. Microsc. Microanal. 25 , 220–221 (2019).

Savitzky, B. H. et al. py4dstem: A software package for multimodal analysis of four-dimensional scanning transmission electron microscopy datasets. arXiv. https://arxiv.org/abs/2003.09523 (2020).

Madsen, J. & Susi, T. The abtem code: transmission electron microscopy from first principles. Open Res. Euro. 1 , 24 (2021).

Koch, C. T. Determination of core structure periodicity and point defect density along dislocations . (Arizona State University, 2002).

Allen, L. J. et al. Modelling the inelastic scattering of fast electrons. Ultramicroscopy 151 , 11–22 (2015).

Maxim, Z., Jesse, S., Sumpter, B. G., Kalinin, S. V. & Dyck, O. Tracking atomic structure evolution during directed electron beam induced si-atom motion in graphene via deep machine learning. Nanotechnology 32 , 035703 (2020).

Khadangi, A., Boudier, T. & Rajagopal, V. Em-net: Deep learning for electron microscopy image segmentation . in 2020 25th International Conference on Pattern Recognition (ICPR) , 31–38 (IEEE, 2021).

Meyer, C. et al. Nion swift: Open source image processing software for instrument control, data acquisition, organization, visualization, and analysis using python. Microsc. Microanal. 25 , 122–123 (2019).

Kim, J., Tiong, L. C. O., Kim, D. & Han, S. S. Deep learning-based prediction of material properties using chemical compositions and diffraction patterns as experimentally accessible inputs. J. Phys. Chem Lett. 12 , 8376–8383 (2021).

Von Chamier, L. et al. Zerocostdl4mic: an open platform to simplify access and use of deep-learning in microscopy. BioRxiv. https://www.biorxiv.org/content/10.1101/2020.03.20.000133v4 (2020).

Jha, D. et al. Peak area detection network for directly learning phase regions from raw x-ray diffraction patterns . in 2019 International Joint Conference on Neural Networks (IJCNN) , 1–8 (IEEE, 2019).

Hawizy, L., Jessop, D. M., Adams, N. & Murray-Rust, P. Chemicaltagger: A tool for semantic text-mining in chemistry. J. Cheminformatics 3 , 1–13 (2011).

Corbett, P. & Boyle, J. Chemlistem: chemical named entity recognition using recurrent neural networks. J. Cheminformatics 10 , 1–9 (2018).

Rocktäschel, T., Weidlich, M. & Leser, U. Chemspot: a hybrid system for chemical named entity recognition. Bioinformatics 28 , 1633–1640 (2012).

Jessop, D. M., Adams, S. E., Willighagen, E. L., Hawizy, L. & Murray-Rust, P. Oscar4: a flexible architecture for chemical text-mining. J. Cheminformatics 3 , 1–12 (2011).

Leaman, R., Wei, C.-H. & Lu, Z. tmchem: a high performance approach for chemical named entity recognition and normalization. J. Cheminformatics 7 , 1–10 (2015).

Suzuki, Y. et al. Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach. Sci. Rep. 10 , 21790 (2020).

Download references

Acknowledgements

Contributions from K.C. were supported by the financial assistance award 70NANB19H117 from the U.S. Department of Commerce, National Institute of Standards and Technology. E.A.H. and R.C. (CMU) were supported by the National Science Foundation under grant CMMI-1826218 and the Air Force D3OM2S Center of Excellence under agreement FA8650-19-2-5209. A.J., C.C., and S.P.O. were supported by the Materials Project, funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division under contract no. DE-AC02-05-CH11231: Materials Project program KC23MP. S.J.L.B. was supported by the U.S. National Science Foundation through grant DMREF-1922234. A.A. and A.C. were supported by NIST award 70NANB19H005 and NSF award CMMI-2053929.

Author information

Authors and affiliations.

Materials Science and Engineering Division, National Institute of Standards and Technology, Gaithersburg, MD, 20899, USA

Kamal Choudhary & Francesca Tavazza

Theiss Research, La Jolla, CA, 92037, USA

Kamal Choudhary

DeepMaterials LLC, Silver Spring, MD, 20906, USA

Material Measurement Science Division, National Institute of Standards and Technology, Gaithersburg, MD, 20899, USA

Brian DeCost

Department of NanoEngineering, University of California San Diego, San Diego, CA, 92093, USA

Chi Chen & Shyue Ping Ong

Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, CA, USA

• Anubhav Jain

Department of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, PA, 15213, USA

Ryan Cohn & Elizabeth Holm

Department of Materials Science and Engineering, Northwestern University, Evanston, IL, 60208, USA

Cheol Woo Park & Chris Wolverton

Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, 60208, USA

Alok Choudhary & Ankit Agrawal

Department of Applied Physics and Applied Mathematics and the Data Science Institute, Fu Foundation School of Engineering and Applied Sciences, Columbia University, New York, NY, 10027, USA

Simon J. L. Billinge

You can also search for this author in PubMed   Google Scholar

Contributions

The authors contributed equally to the search as well as analysis of the literature and writing of the manuscript.

Corresponding author

Correspondence to Kamal Choudhary .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Choudhary, K., DeCost, B., Chen, C. et al. Recent advances and applications of deep learning methods in materials science. npj Comput Mater 8 , 59 (2022). https://doi.org/10.1038/s41524-022-00734-6

Download citation

Received : 25 October 2021

Accepted : 24 February 2022

Published : 05 April 2022

DOI : https://doi.org/10.1038/s41524-022-00734-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Analysis of solar energy potentials of five selected south-east cities in nigeria using deep learning algorithms.

• Samuel Ikemba
• Kim Song-hyun
• Akeeb Adepoju Fawole

Sustainable Energy Research (2024)

Structured information extraction from scientific text with large language models

• John Dagdelen
• Alexander Dunn

Nature Communications (2024)

Methods and applications of machine learning in computational design of optoelectronic semiconductors

• Xiaoyu Yang
• Lijun Zhang

Science China Materials (2024)

Data-driven analysis of spinodoid topologies: anisotropy, inverse design, and elasticity tensor distribution

• Farshid Golnary
• Mohsen Asghari

International Journal of Mechanics and Materials in Design (2024)

Microstructural Feature Extraction by a Convolutional Neural Network for Cold Spray of Aluminum Alloys

Journal of Thermal Spray Technology (2024)

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Analytics Insight

Top 10 Research Challenge Areas to Pursue in Data Science

Disclaimer: Any financial and crypto market information given on Analytics Insight are sponsored articles, written for informational purpose only and is not an investment advice. The readers are further advised that Crypto products and NFTs are unregulated and can be highly risky. There may be no regulatory recourse for any loss from such transactions. Conduct your own research by contacting financial experts before making any investment decisions. The decision to read hereinafter is purely a matter of choice and shall be construed as an express undertaking/guarantee in favour of Analytics Insight of being absolved from any/ all potential legal action, or enforceable claims. We do not represent nor own any cryptocurrency, any complaints, abuse or concerns with regards to the information provided shall be immediately informed here .

You May Also Like

Best Crypto for the Bull Run: Scorpion Casino Outshines the Resurgent Cosmos and Jupiter Duo

Strategies to Earn US$1000/Day Using Devin AI

Bitcoin Might Just Survive these Last Two Months of 2022!

InQubeta Gains Traction Over Popular Meme Coins PEPE and FLOKI

Analytics Insight® is an influential platform dedicated to insights, trends, and opinion from the world of data-driven technologies. It monitors developments, recognition, and achievements made by Artificial Intelligence, Big Data and Analytics companies across the globe.

• Select Language:
• Privacy Policy
• Content Licensing
• Terms & Conditions
• Submit an Interview

Special Editions

• Dec – Crypto Weekly Vol-1
• 40 Under 40 Innovators
• Women In Technology
• Market Reports
• AI Glossary
• Infographics

Latest Issue

Disclaimer: Any financial and crypto market information given on Analytics Insight is written for informational purpose only and is not an investment advice. Conduct your own research by contacting financial experts before making any investment decisions, more information here .

Second Menu

• Getting Published
• Open Research
• Communicating Research
• Life in Research
• For Editors
• For Peer Reviewers
• Research Integrity

Data Science and Engineering: Research Areas

Author: guest contributor.

Data science has emerged as an independent domain in the decade starting 2010 with the explosive growth in big data analytics, cloud, and IoT technology capabilities. A data scientist requires fundamental knowledge in the areas of computer science, statistics, and machine learning, which he may use to solve problems in a variety of domains. We may define data science as a study of scientific principles that describe data and their inter-relationship. Some of the current areas of research in Data Science and Engineering are categorized and enumerated below : 

1. Artificial Intelligence / Machine Learning :

While human beings learn from experience, machines learn from data and improve their accuracy over time. AI applications attempt to mimic human intelligence by a computer, robot, or other machines. AI/ML has brought disruptive innovations in business and social life. One of the emerging areas in AI is generative artificial intelligence algorithms that use reinforcement learning for content creation such as text, code, audio, images, and videos. The AI based chatbot ‘ChatGPT’ from Open AI is a product in this line. ChatGPT can code computer programs, compose music, write short stories and essays, and much more!

2. Automation: 

Some of the research areas in automation include public ride-share services (e.g., uber platform), self-driving vehicles, and automation of the manufacturing industry. AI/ML techniques are widely used in industries for the identification of unusual patterns in sensor readings from machinery and equipment for the detection or prevention of malfunction.

3. Business:

As we know, social media provide opportunities for people to interact, share, and participate in numerous activities in a massive way. A marketing researcher may analyze this data to gain an understanding of human sentiments and behavior unobtrusively, at a scale unheard of in traditional marketing. We come across personalized product recommender systems almost every day. Content-based recommender systems guess user’s intentions based on the history of their previous activities. Collaborative recommender systems use data mining techniques to make personalized product recommendations, during live customer transactions, based on the opinions of customers with similar profile. 

Data science finds numerous applications in finance like stock market analysis; targeted marketing; and detection of unusual transaction patterns, fraudulent credit card transactions, and money laundering. Financial markets are complex and chaotic. However, AI technologies make it possible to process massive amounts of real-time data, leading to accurate forecast and trade. Stock Hero, Scanz, Tickeron, Impertive execution, and Algoriz are some of the AI based products for stock market prediction. 

4. Computer Vision and NLP:

AI/ML models are extensively used in digital image processing, computer vision, speech recognition, and natural language processing (NLP). In image processing, we use mathematical transformations to enhance an image. These transformations typically include smoothing, sharpening, contrasting, and stretching. From the transformed images we can extract various types of features - edges, corners, ridges, and blobs/regions. The objective of computer vision is to identify objects (or images). To achieve this, the input image is processed, features are extracted, and using the features the object is classified (or identified).

Natural language processing techniques are used to understand human language in written or spoken form and translate it to another language or respond to commands. Voice-operated GPS systems, translation tools, speech-to-text dictation, and customer service chatbots are all applications of NLP. Siri, and Alexa are popular NLP products. 

5. Data Mining

Data mining is the process of cleaning and analyzing data to identify hidden patterns and trends that are not readily discernible from a conventional spread sheet. Building models for classification and clustering in high dimensional, streaming, and/or big data space is an area that receives much attention from researchers. Network-graph based algorithms are being developed for representing and analyzing the interactions in social media such as facebook, twitter, linkedin, instagram, and web sites. 

6. Data Management:

Information storage and retrieval is area that is concerned with effective and efficient storage and retrieval of digital documents in multiple data formats, using their semantic content. Government regulations and individual privacy concerns necessitate cryptographic methods for storing and sharing data such as secure multi-party computation, homomorphic encryption, and differential privacy. 

Data-stream processing needs specialized algorithms and techniques for doing computations on huge data that arrive fast and require immediate processing – e.g., satellite images, data from sensors, internet traffic, and web searches. Some of the other areas of research in data management include big data databases, cloud computing architectures, crowd sourcing, human-machine interaction, and data governance. 

7. Data visualization

Visualizing complex, big, and / or streaming data, such as the onset of a storm or a cosmic event, demands advanced techniques. In data visualization, the user usually follows a three-step process - get an overview of the data, identify interesting patterns, and drill-down for final details. In most cases, the input data is subjected to mathematical transformations and statistical summarizations. The visualization of the real physical world may be further enhanced using audio-visual techniques or other sensory stimuli delivered by technology. This technique is called augmented reality. Virtual reality provides a computer-generated virtual environment giving an immersive experience to the users. For example, ‘Pokémon GO’ that allows you play the game Pokémon is an AR product released in 2016; Google Earth VR is VR product that ‘puts the whole world within your reach’.

8. Genetic Studies:

Genetic studies are path breaking investigation of the biological basis of inherited and acquired genetic variation using advanced statistical methods. The human genome project (1990 – 2003) produced a genome sequence that accounted for over 90% of the human genome. The project cost was about USD 3 billion. The data underlying a single human genome sequence is about 200 gigabytes. The digital revolution has made astounding possibilities to pinpoint human evolution with marked accuracy. Note that the cost of sequencing the entire genome of a human cell has fallen from USD 100,000,000 in the year 2000 to USD 800 in 2020! 

9. Government:

Governments need smart and effective platforms for interacting with citizens, data collection, validation, and analysis. Data driven tools and AI/ML techniques are used for fighting terrorism, intervention in street crimes, and tackling cyber-attack. Data science also provides support in rendering public services, national and social security, and emergency responses.

10. Healthcare:

The most important contribution of data science in the pharmaceutical industry is to provide computational support for cost effective drug discovery using AI/ML techniques. AI/ML supports medical diagnosis, preventive care, and prediction of failures based on historical data. Study of genetic data helps in the identification of anomalies, prediction of possible failures and personalized drug suggestions, e.g., in cancer treatment. Medical image processing use data science techniques to visualize, interrogate, identify, and treat deformities in the internal organs and systems.

Electronic health records (EHR) are concerned with the storage of data arriving in multiple formats, data privacy (e.g., conformance with HIPAA privacy regulations), and data sharing between stakeholders. Wearable technology provides electronic devices and platforms for collecting and analyzing data related to personal health and exercise – for example, Fitbit and smartwatches. The Covid-19 pandemic demonstrated the power of data science in monitoring and controlling an epidemic as well as developing drugs in record time. 

11. Responsible AI : 

AI systems support complex decision making in various domains such as autonomous vehicles, healthcare, public safety, HR practices etc. To trust the AI systems, their decisions must be reliable, explainable, accountable, and ethical. There is ongoing research on how these facets can be built into AI algorithms.

This book appears in the book series Transactions on Computer Systems and Networks . If you are interested in writing a book in the series,  then please click here to complete and submit the relevant form.

Srikrishnan Sundararajan, PhD in Computer Applications, is a retired senior professor of business analytics, Loyola institute of business administration, Chennai, India. He has held various tenured and visiting professorships in Business Analytics, and Computer Science for over 10 years. He has 25 years of experience as a consultant in the information technology industry in India and the USA, in information systems development and technology support. 

He is the author of the forthcoming book ‘Multivariate Analysis and Machine Learning Techniques - Feature Analysis in Data Science using Python’ published by Springer Nature (ISBN.9789819903528). This book offers a comprehensive first-level introduction to data science including python programming, probability and statistics, multivariate analysis, survival analysis, AI/ML, and other computational techniques.

Guest Contributors include Springer Nature staff and authors, industry experts, society partners, and many others. If you are interested in being a Guest Contributor, please contact us via email: [email protected] .

• Tools & Services
• Account Development
• Sales and account contacts
• Professional
• Press office
• Locations & Contact

We are a world leading research, educational and professional publisher. Visit our main website for more information.

• © 2023 Springer Nature
• General terms and conditions
• Your US State Privacy Rights
• Your Privacy Choices / Manage Cookies
• Accessibility
• Legal notice
• Help us to improve this site, send feedback.

• Data, AI, & Machine Learning
• Managing Technology
• Social Responsibility
• Workplace, Teams, & Culture
• AI & Machine Learning
• Diversity & Inclusion
• Big ideas Research Projects
• Artificial Intelligence and Business Strategy
• Responsible AI
• Future of the Workforce
• Future of Leadership
• All Research Projects

AI in Action

• Most Popular
• The Truth Behind the Nursing Crisis
• Work/23: The Big Shift
• Coaching for the Future-Forward Leader
• Measuring Culture

The spring 2024 issue’s special report looks at how to take advantage of market opportunities in the digital space, and provides advice on building culture and friendships at work; maximizing the benefits of LLMs, corporate venture capital initiatives, and innovation contests; and scaling automation and digital health platform.

• Past Issues
• Upcoming Events
• Video Archive
• Me, Myself, and AI
• Three Big Points

Five Key Trends in AI and Data Science for 2024

These developing issues should be on every leader’s radar screen, data executives say.

• Data, AI, & Machine Learning
• AI & Machine Learning
• Data & Data Culture
• Technology Implementation

Carolyn Geason-Beissel/MIT SMR | Getty Images

Artificial intelligence and data science became front-page news in 2023. The rise of generative AI, of course, drove this dramatic surge in visibility. So, what might happen in the field in 2024 that will keep it on the front page? And how will these trends really affect businesses?

During the past several months, we’ve conducted three surveys of data and technology executives. Two involved MIT’s Chief Data Officer and Information Quality Symposium attendees — one sponsored by Amazon Web Services (AWS) and another by Thoughtworks . The third survey was conducted by Wavestone , formerly NewVantage Partners, whose annual surveys we’ve written about in the past . In total, the new surveys involved more than 500 senior executives, perhaps with some overlap in participation.

Get Updates on Leading With AI and Data

Get monthly insights on how artificial intelligence impacts your organization and what it means for your company and customers.

Please enter a valid email address

Thank you for signing up

Privacy Policy

Surveys don’t predict the future, but they do suggest what those people closest to companies’ data science and AI strategies and projects are thinking and doing. According to those data executives, here are the top five developing issues that deserve your close attention:

1. Generative AI sparkles but needs to deliver value.

As we noted, generative AI has captured a massive amount of business and consumer attention. But is it really delivering economic value to the organizations that adopt it? The survey results suggest that although excitement about the technology is very high , value has largely not yet been delivered. Large percentages of respondents believe that generative AI has the potential to be transformational; 80% of respondents to the AWS survey said they believe it will transform their organizations, and 64% in the Wavestone survey said it is the most transformational technology in a generation. A large majority of survey takers are also increasing investment in the technology. However, most companies are still just experimenting, either at the individual or departmental level. Only 6% of companies in the AWS survey had any production application of generative AI, and only 5% in the Wavestone survey had any production deployment at scale.

Surveys suggest that though excitement about generative AI is very high, value has largely not yet been delivered.

Production deployments of generative AI will, of course, require more investment and organizational change, not just experiments. Business processes will need to be redesigned, and employees will need to be reskilled (or, probably in only a few cases, replaced by generative AI systems). The new AI capabilities will need to be integrated into the existing technology infrastructure.

Perhaps the most important change will involve data — curating unstructured content, improving data quality, and integrating diverse sources. In the AWS survey, 93% of respondents agreed that data strategy is critical to getting value from generative AI, but 57% had made no changes to their data thus far.

2. Data science is shifting from artisanal to industrial.

Companies feel the need to accelerate the production of data science models . What was once an artisanal activity is becoming more industrialized. Companies are investing in platforms, processes and methodologies, feature stores, machine learning operations (MLOps) systems, and other tools to increase productivity and deployment rates. MLOps systems monitor the status of machine learning models and detect whether they are still predicting accurately. If they’re not, the models might need to be retrained with new data.

Producing data models — once an artisanal activity — is becoming more industrialized.

Most of these capabilities come from external vendors, but some organizations are now developing their own platforms. Although automation (including automated machine learning tools, which we discuss below) is helping to increase productivity and enable broader data science participation, the greatest boon to data science productivity is probably the reuse of existing data sets, features or variables, and even entire models.

3. Two versions of data products will dominate.

In the Thoughtworks survey, 80% of data and technology leaders said that their organizations were using or considering the use of data products and data product management. By data product , we mean packaging data, analytics, and AI in a software product offering, for internal or external customers. It’s managed from conception to deployment (and ongoing improvement) by data product managers. Examples of data products include recommendation systems that guide customers on what products to buy next and pricing optimization systems for sales teams.

But organizations view data products in two different ways. Just under half (48%) of respondents said that they include analytics and AI capabilities in the concept of data products. Some 30% view analytics and AI as separate from data products and presumably reserve that term for reusable data assets alone. Just 16% say they don’t think of analytics and AI in a product context at all.

We have a slight preference for a definition of data products that includes analytics and AI, since that is the way data is made useful. But all that really matters is that an organization is consistent in how it defines and discusses data products. If an organization prefers a combination of “data products” and “analytics and AI products,” that can work well too, and that definition preserves many of the positive aspects of product management. But without clarity on the definition, organizations could become confused about just what product developers are supposed to deliver.

4. Data scientists will become less sexy.

Data scientists, who have been called “ unicorns ” and the holders of the “ sexiest job of the 21st century ” because of their ability to make all aspects of data science projects successful, have seen their star power recede. A number of changes in data science are producing alternative approaches to managing important pieces of the work. One such change is the proliferation of related roles that can address pieces of the data science problem. This expanding set of professionals includes data engineers to wrangle data, machine learning engineers to scale and integrate the models, translators and connectors to work with business stakeholders, and data product managers to oversee the entire initiative.

Another factor reducing the demand for professional data scientists is the rise of citizen data science , wherein quantitatively savvy businesspeople create models or algorithms themselves. These individuals can use AutoML, or automated machine learning tools, to do much of the heavy lifting. Even more helpful to citizens is the modeling capability available in ChatGPT called Advanced Data Analysis . With a very short prompt and an uploaded data set, it can handle virtually every stage of the model creation process and explain its actions.

Of course, there are still many aspects of data science that do require professional data scientists. Developing entirely new algorithms or interpreting how complex models work, for example, are tasks that haven’t gone away. The role will still be necessary but perhaps not as much as it was previously — and without the same degree of power and shimmer.

5. Data, analytics, and AI leaders are becoming less independent.

This past year, we began to notice that increasing numbers of organizations were cutting back on the proliferation of technology and data “chiefs,” including chief data and analytics officers (and sometimes chief AI officers). That CDO/CDAO role, while becoming more common in companies, has long been characterized by short tenures and confusion about the responsibilities. We’re not seeing the functions performed by data and analytics executives go away; rather, they’re increasingly being subsumed within a broader set of technology, data, and digital transformation functions managed by a “supertech leader” who usually reports to the CEO. Titles for this role include chief information officer, chief information and technology officer, and chief digital and technology officer; real-world examples include Sastry Durvasula at TIAA, Sean McCormack at First Group, and Mojgan Lefebvre at Travelers.

Related Articles

This evolution in C-suite roles was a primary focus of the Thoughtworks survey, and 87% of respondents (primarily data leaders but some technology executives as well) agreed that people in their organizations are either completely, to a large degree, or somewhat confused about where to turn for data- and technology-oriented services and issues. Many C-level executives said that collaboration with other tech-oriented leaders within their own organizations is relatively low, and 79% agreed that their organization had been hindered in the past by a lack of collaboration.

We believe that in 2024, we’ll see more of these overarching tech leaders who have all the capabilities to create value from the data and technology professionals reporting to them. They’ll still have to emphasize analytics and AI because that’s how organizations make sense of data and create value with it for employees and customers. Most importantly, these leaders will need to be highly business-oriented, able to debate strategy with their senior management colleagues, and able to translate it into systems and insights that make that strategy a reality.

About the Authors

Thomas H. Davenport ( @tdav ) is the President’s Distinguished Professor of Information Technology and Management at Babson College, a fellow of the MIT Initiative on the Digital Economy, and senior adviser to the Deloitte Chief Data and Analytics Officer Program. He is coauthor of All in on AI: How Smart Companies Win Big With Artificial Intelligence (HBR Press, 2023) and Working With AI: Real Stories of Human-Machine Collaboration (MIT Press, 2022). Randy Bean ( @randybeannvp ) is an industry thought leader, author, founder, and CEO and currently serves as innovation fellow, data strategy, for global consultancy Wavestone. He is the author of Fail Fast, Learn Faster: Lessons in Data-Driven Leadership in an Age of Disruption, Big Data, and AI (Wiley, 2021).

More Like This

Add a comment cancel reply.

You must sign in to post a comment. First time here? Sign up for a free account : Comment on articles and get access to many more articles.

Comment (1)

Nicolas corzo.

99+ Interesting Data Science Research Topics For Students In 2024

In today’s information-driven world, data science research stands as a pivotal domain shaping our understanding and application of vast data sets. It amalgamates statistics, computer science, and domain knowledge to extract valuable insights from data. Understanding ‘What Is Data Science?’ is fundamental—a field exploring patterns, trends, and solutions embedded within data.

However, the significance of data science research papers in a student’s life cannot be overstated. They foster critical thinking, analytical skills, and a deeper comprehension of the subject matter. To aid students in navigating this realm effectively, this blog dives into essential elements integral to a data science research paper, while also offering a goldmine of 99+ engaging and timely data science research topics for 2024.

Unraveling tips for crafting an impactful research paper and insights on choosing the right topic, this blog is a compass for students exploring data science research topics. Stay tuned to unearth more about ‘data science research topics’ and refine your academic journey.

What Is Data Science?

Table of Contents

Data Science is like a detective for information! It’s all about uncovering secrets and finding valuable stuff in heaps of data. Imagine you have a giant puzzle with tons of pieces scattered around. Data Science helps in sorting these pieces and figuring out the picture they create. It uses tools and skills from math, computer science, and knowledge about different fields to solve real-world problems.

In simpler terms, Data Science is like a chef in a kitchen, blending ingredients to create a perfect dish. Instead of food, it combines data—numbers, words, pictures—to cook up solutions. It helps in understanding patterns, making predictions, and answering tricky questions by exploring data from various sources. In essence, Data Science is the magic that turns data chaos into meaningful insights that can guide decisions and make life better.

Importance Of Data Science Research Paper In Student’s Life

Data Science research papers are like treasure maps for students! They’re super important because they teach students how to explore and understand the world of data. Writing these papers helps students develop problem-solving skills, think critically, and become better at analyzing information. It’s like a fun adventure where they learn how to dig into data and uncover valuable insights that can solve real-world problems.

• Enhances critical thinking: Research papers challenge students to analyze and interpret data critically, honing their thinking skills.
• Fosters analytical abilities: Students learn to sift through vast amounts of data, extracting meaningful patterns and information.
• Encourages exploration: Engaging in research encourages students to explore diverse data sources, broadening their knowledge horizon.
• Develops communication skills: Writing research papers hones students’ ability to articulate complex findings and ideas clearly.
• Prepares for real-world challenges: Through research, students learn to apply theoretical knowledge to practical problems, preparing them for future endeavors.

Elements That Must Be Present In Data Science Research Paper

Here are some elements that must be present in data science research paper:

1. Clear Objective

A data science research paper should start with a clear goal, stating what the study aims to investigate or achieve. This objective guides the entire paper, helping readers understand the purpose and direction of the research.

2. Detailed Methodology

Explaining how the research was conducted is crucial. The paper should outline the tools, techniques, and steps used to collect, analyze, and interpret data. This section allows others to replicate the study and validate its findings.

3. Accurate Data Presentation

Presenting data in an organized and understandable manner is key. Graphs, charts, and tables should be used to illustrate findings clearly, aiding readers’ comprehension of the results.

4. Thorough Analysis and Interpretation

Simply presenting data isn’t enough; the paper should delve into a deep analysis, explaining the meaning behind the numbers. Interpretation helps draw conclusions and insights from the data.

5. Conclusive Findings and Recommendations

A strong conclusion summarizes the key findings of the research. It should also offer suggestions or recommendations based on the study’s outcomes, indicating potential avenues for future exploration.

Here are some interesting data science research topics for students in 2024:

Natural Language Processing (NLP)

• Multi-modal Contextual Understanding: Integrating text, images, and audio to enhance NLP models’ comprehension abilities.
• Cross-lingual Transfer Learning: Investigating methods to transfer knowledge from one language to another for improved translation and understanding.
• Emotion Detection in Text: Developing models to accurately detect and interpret emotions conveyed in textual content.
• Sarcasm Detection in Social Media: Building algorithms that can identify and understand sarcastic remarks in online conversations.
• Language Generation for Code: Generating code snippets and scripts from natural language descriptions using NLP techniques.
• Bias Mitigation in Language Models: Developing strategies to mitigate biases present in large language models and ensure fairness in generated content.
• Dialogue Systems for Personalized Assistance: Creating intelligent conversational agents that provide personalized assistance based on user preferences and history.
• Summarization of Legal Documents: Developing NLP models capable of summarizing lengthy legal documents for quick understanding and analysis.
• Understanding Contextual Nuances in Sentiment Analysis: Enhancing sentiment analysis models to better comprehend contextual nuances and sarcasm in text.
• Hate Speech Detection and Moderation: Building systems to detect and moderate hate speech and offensive language in online content.

Computer Vision

• Weakly Supervised Object Detection: Exploring methods to train object detection models with limited annotated data.
• Video Action Recognition in Uncontrolled Environments: Developing models that can recognize human actions in videos captured in uncontrolled settings.
• Image Generation and Translation: Investigating techniques to generate realistic images from textual descriptions and translate images across different domains.
• Scene Understanding in Autonomous Systems: Enhancing computer vision algorithms for better scene understanding in autonomous vehicles and robotics.
• Fine-grained Visual Classification: Improving models to classify objects at a more granular level, distinguishing subtle differences within similar categories.
• Visual Question Answering (VQA): Creating systems capable of answering questions based on visual input, requiring reasoning and comprehension abilities.
• Medical Image Analysis for Disease Diagnosis: Developing computer vision models for accurate and early diagnosis of diseases from medical images.
• Action Localization in Videos: Building models to precisely localize and recognize specific actions within video sequences.
• Image Captioning with Contextual Understanding: Generating captions for images considering the context and relationships between objects.
• Human Pose Estimation in Real-time: Improving algorithms for real-time estimation of human poses in videos for applications like motion analysis and gaming.

Machine Learning Algorithms

• Self-supervised Learning Techniques: Exploring novel methods for training machine learning models without explicit supervision.
• Continual Learning in Dynamic Environments: Investigating algorithms that can continuously learn and adapt to changing data distributions and tasks.
• Explainable AI for Model Interpretability: Developing techniques to explain the decisions and predictions made by complex machine learning models.
• Transfer Learning for Small Datasets: Techniques to effectively transfer knowledge from large datasets to small or domain-specific datasets.
• Adaptive Learning Rate Optimization: Enhancing optimization algorithms to dynamically adjust learning rates based on data characteristics.
• Robustness to Adversarial Attacks: Building models resistant to adversarial attacks, ensuring stability and security in machine learning applications.
• Active Learning Strategies: Investigating methods to select and label the most informative data points for model training to minimize labeling efforts.
• Privacy-preserving Machine Learning: Developing algorithms that can train models on sensitive data while preserving individual privacy.
• Fairness-aware Machine Learning: Techniques to ensure fairness and mitigate biases in machine learning models across different demographics.
• Multi-task Learning for Jointly Learning Tasks: Exploring approaches to jointly train models on multiple related tasks to improve overall performance.

Deep Learning

• Graph Neural Networks for Representation Learning: Using deep learning techniques to learn representations from graph-structured data.
• Transformer Models for Image Processing: Adapting transformer architectures for image-related tasks, such as image classification and generation.
• Few-shot Learning Strategies: Investigating methods to enable deep learning models to learn from a few examples in new categories.
• Memory-Augmented Neural Networks: Enhancing neural networks with external memory for improved learning and reasoning capabilities.
• Neural Architecture Search (NAS): Automating the design of neural network architectures for specific tasks or constraints.
• Meta-learning for Fast Adaptation: Developing models capable of quickly adapting to new tasks or domains with minimal data.
• Deep Reinforcement Learning for Robotics: Utilizing deep RL techniques for training robots to perform complex tasks in real-world environments.
• Generative Adversarial Networks (GANs) for Data Augmentation: Using GANs to generate synthetic data for enhancing training datasets.
• Variational Autoencoders for Unsupervised Learning: Exploring VAEs for learning latent representations of data without explicit supervision.
• Lifelong Learning in Deep Networks: Strategies to enable deep networks to continually learn from new data while retaining past knowledge.

Big Data Analytics

• Streaming Data Analysis for Real-time Insights: Techniques to analyze and derive insights from continuous streams of data in real-time.
• Scalable Algorithms for Massive Graphs: Developing algorithms that can efficiently process and analyze large-scale graph-structured data.
• Anomaly Detection in High-dimensional Data: Detecting anomalies and outliers in high-dimensional datasets using advanced statistical methods and machine learning.
• Personalization and Recommendation Systems: Enhancing recommendation algorithms for providing personalized and relevant suggestions to users.
• Data Quality Assessment and Improvement: Methods to assess, clean, and enhance the quality of big data to improve analysis and decision-making.
• Time-to-Event Prediction in Time-series Data: Predicting future events or occurrences based on historical time-series data.
• Geospatial Data Analysis and Visualization: Analyzing and visualizing large-scale geospatial data for various applications such as urban planning, disaster management, etc.
• Privacy-preserving Big Data Analytics: Ensuring data privacy while performing analytics on large-scale datasets in distributed environments.
• Graph-based Deep Learning for Network Analysis: Leveraging deep learning techniques for network analysis and community detection in large-scale networks.
• Dynamic Data Compression Techniques: Developing methods to compress and store large volumes of data efficiently without losing critical information.

Healthcare Analytics

• Predictive Modeling for Patient Outcomes: Using machine learning to predict patient outcomes and personalize treatments based on individual health data.
• Clinical Natural Language Processing for Electronic Health Records (EHR): Extracting valuable information from unstructured EHR data to improve healthcare delivery.
• Wearable Devices and Health Monitoring: Analyzing data from wearable devices to monitor and predict health conditions in real-time.
• Drug Discovery and Development using AI: Utilizing machine learning and AI for efficient drug discovery and development processes.
• Predictive Maintenance in Healthcare Equipment: Developing models to predict and prevent equipment failures in healthcare settings.
• Disease Clustering and Stratification: Grouping diseases based on similarities in symptoms, genetic markers, and response to treatments.
• Telemedicine Analytics: Analyzing data from telemedicine platforms to improve remote healthcare delivery and patient outcomes.
• AI-driven Radiomics for Medical Imaging: Using AI to extract quantitative features from medical images for improved diagnosis and treatment planning.
• Healthcare Resource Optimization: Optimizing resource allocation in healthcare facilities using predictive analytics and operational research techniques.
• Patient Journey Analysis and Personalized Care Pathways: Analyzing patient trajectories to create personalized care pathways and improve healthcare outcomes.

Time Series Analysis

• Forecasting Volatility in Financial Markets: Predicting and modeling volatility in stock prices and financial markets using time series analysis.
• Dynamic Time Warping for Similarity Analysis: Utilizing DTW to measure similarities between time series data, especially in scenarios with temporal distortions.
• Seasonal Pattern Detection and Analysis: Identifying and modeling seasonal patterns in time series data for better forecasting.
• Time Series Anomaly Detection in Industrial IoT: Detecting anomalies in industrial sensor data streams to prevent equipment failures and improve maintenance.
• Multivariate Time Series Forecasting: Developing models to forecast multiple related time series simultaneously, considering interdependencies.
• Non-linear Time Series Analysis Techniques: Exploring non-linear models and methods for analyzing complex time series data.
• Time Series Data Compression for Efficient Storage: Techniques to compress and store time series data efficiently without losing crucial information.
• Event Detection and Classification in Time Series: Identifying and categorizing specific events or patterns within time series data.
• Time Series Forecasting with Uncertainty Estimation: Incorporating uncertainty estimation into time series forecasting models for better decision-making.
• Dynamic Time Series Graphs for Network Analysis: Representing and analyzing dynamic relationships between entities over time using time series graphs.

Reinforcement Learning

• Multi-agent Reinforcement Learning for Collaboration: Developing strategies for multiple agents to collaborate and solve complex tasks together.
• Hierarchical Reinforcement Learning: Utilizing hierarchical structures in RL for solving tasks with varying levels of abstraction and complexity.
• Model-based Reinforcement Learning for Sample Efficiency: Incorporating learned models into RL for efficient exploration and planning.
• Robotic Manipulation with Reinforcement Learning: Training robots to perform dexterous manipulation tasks using RL algorithms.
• Safe Reinforcement Learning: Ensuring that RL agents operate safely and ethically in real-world environments, minimizing risks.
• Transfer Learning in Reinforcement Learning: Transferring knowledge from previously learned tasks to expedite learning in new but related tasks.
• Curriculum Learning Strategies in RL: Designing learning curricula to gradually expose RL agents to increasingly complex tasks.
• Continuous Control in Reinforcement Learning: Exploring techniques for continuous control tasks that require precise actions in a continuous action space.
• Reinforcement Learning for Adaptive Personalization: Utilizing RL to personalize experiences or recommendations for individuals in dynamic environments.
• Reinforcement Learning in Healthcare Decision-making: Using RL to optimize treatment strategies and decision-making in healthcare settings.

Data Mining

• Graph Mining for Social Network Analysis: Extracting valuable insights from social network data using graph mining techniques.
• Sequential Pattern Mining for Market Basket Analysis: Discovering sequential patterns in customer purchase behaviors for market basket analysis.
• Clustering Algorithms for High-dimensional Data: Developing clustering techniques suitable for high-dimensional datasets.
• Frequent Pattern Mining in Healthcare Datasets: Identifying frequent patterns in healthcare data for actionable insights and decision support.
• Outlier Detection and Fraud Analysis: Detecting anomalies and fraudulent activities in various domains using data mining approaches.
• Opinion Mining and Sentiment Analysis in Reviews: Analyzing opinions and sentiments expressed in product or service reviews to derive insights.
• Data Mining for Personalized Learning: Mining educational data to personalize learning experiences and adapt teaching methods.
• Association Rule Mining in Internet of Things (IoT) Data: Discovering meaningful associations and patterns in IoT-generated data streams.
• Multi-modal Data Fusion for Comprehensive Analysis: Integrating information from multiple data modalities for a holistic understanding and analysis.
• Data Mining for Energy Consumption Patterns: Analyzing energy usage data to identify patterns and optimize energy consumption in various sectors.

Ethical AI and Bias Mitigation

• Fairness Metrics and Evaluation in AI Systems: Developing metrics and evaluation frameworks to assess the fairness of AI models.
• Bias Detection and Mitigation in Facial Recognition: Addressing biases present in facial recognition systems to ensure equitable performance across demographics.
• Algorithmic Transparency and Explainability: Designing methods to make AI algorithms more transparent and understandable to stakeholders.
• Fair Representation Learning in Unbalanced Datasets: Learning fair representations from imbalanced data to reduce biases in downstream tasks.
• Fairness-aware Recommender Systems: Ensuring fairness and reducing biases in recommendation algorithms across diverse user groups.
• Ethical Considerations in AI for Criminal Justice: Investigating the ethical implications of AI-based decision-making in criminal justice systems.
• Debiasing Techniques in Natural Language Processing: Developing methods to mitigate biases in language models and text generation.
• Diversity and Fairness in Hiring Algorithms: Ensuring diversity and fairness in AI-based hiring systems to minimize biases in candidate selection.
• Ethical AI Governance and Policy: Examining the role of governance and policy frameworks in regulating the development and deployment of AI systems.
• AI Accountability and Responsibility: Addressing ethical dilemmas and defining responsibilities concerning AI system behaviors and decision-making processes.

Tips For Writing An Effective Data Science Research Paper

Here are some tips for writing an effective data science research paper:

Tip 1: Thorough Planning and Organization

Begin by planning your research paper carefully. Outline the sections and information you’ll include, ensuring a logical flow from introduction to conclusion. This organized approach makes writing easier and helps maintain coherence in your paper.

Tip 2: Clarity in Writing Style

Use clear and simple language to communicate your ideas. Avoid jargon or complex terms that might confuse readers. Write in a way that is easy to understand, ensuring your message is effectively conveyed.

Tip 3: Precise and Relevant Information

Include only information directly related to your research topic. Ensure the data, explanations, and examples you use are precise and contribute directly to supporting your arguments or findings.

Tip 4: Effective Data Visualization

Utilize graphs, charts, and tables to present your data visually. Visual aids make complex information easier to comprehend and can enhance the overall presentation of your research findings.

Tip 5: Review and Revise

Before submitting your paper, review it thoroughly. Check for any errors in grammar, spelling, or formatting. Revise sections if necessary to ensure clarity and coherence in your writing. Asking someone else to review it can also provide valuable feedback.

• Hospitality Management Research Topics

Things To Remember While Choosing The Data Science Research Topic

When selecting a data science research topic, consider your interests and its relevance to the field. Ensure the topic is neither too broad nor too narrow, striking a balance that allows for in-depth exploration while staying manageable.

• Relevance and Significance: Choose a topic that aligns with current trends or addresses a significant issue in the field of data science.
• Feasibility : Ensure the topic is researchable within the resources and time available. It should be practical and manageable for the scope of your study.
• Your Interest and Passion: Select a topic that genuinely interests you. Your enthusiasm will drive your motivation and engagement throughout the research process.
• Availability of Data: Check if there’s sufficient data available for analysis related to your chosen topic. Accessible and reliable data sources are vital for thorough research.
• Potential Contribution: Consider how your chosen topic can contribute to existing knowledge or fill a gap in the field. Aim for a topic that adds value and insights to the data science domain.

In wrapping up our exploration of data science research topics, we’ve uncovered a world of importance and guidance for students. From defining data science to understanding its impact on student life, identifying essential elements in research papers, offering a multitude of intriguing topics for 2024, to providing tips for crafting effective papers—the journey has been insightful. 

Remembering the significance of topic selection and the key components of a well-structured paper, this voyage emphasizes how data science opens doors to endless opportunities. It’s not just a subject; it’s the compass guiding tomorrow’s discoveries and innovations in our digital landscape.

Related Posts

Step by Step Guide on The Best Way to Finance Car

The Best Way on How to Get Fund For Business to Grow it Efficiently

• Frontiers in Research Metrics and Analytics
• Scholarly Communication
• Research Topics

Data Science and Artificial Intelligence for (Better) Science

Total Downloads

Total Views and Downloads

About this Research Topic

Data science and Artificial Intelligence (AI) are changing the way we do science. AI is increasingly used in scientific practices, from data discovery to data analysis, for extracting new knowledge out of research artifacts, generating novel and experimentally testable scientific hypotheses, writing, ...

Keywords : Data Science, Artificial Intelligence, Open Data, research life cycle, knowledge production

Important Note : All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

Topic Editors

Topic coordinators, recent articles, submission deadlines.

Submission closed.

Participating Journals

Total views.

• Demographics

No records found

total views article views downloads topic views

Top countries

Top referring sites, about frontiers research topics.

With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.

Data Science Trending Topics

Learn what’s trending in data science. Read the top stories of how data transforms industries like tech, finance, marketing, healthcare, astronomy, natural disaster relief, and more.

Trending Topics

Featured Articles

The Evolution of ChatGPT: History and Future

ChatGPT conquered the world overnight but wasn’t built in a day. How was it created and how has it evolved? Learn about ChatGPT’s history and future.

ChatGPT Code Interpreter: What It Is and How It Works

Data Stories at ZeClinics: Data Science in Biotechnology and the Drug Discovery Industry

Data Analytics for SaaS Companies: A Case Study of Sage

Start Your Data Science Career

Gain the theory and practical skills you need to break into data science.

Join 2M+ students

All Articles

Data Stories at Jayride: Data Science in the Travel Industry

A Guide to Data-Driven Recruitment: Using Workwolf to Reduce Bias and Increase Efficiency

The Aftermath of the Big Tech Layoffs: Who Found a Job and Where?

Who Was Affected by the 2022–2023 Tech Layoffs? A Comprehensive Study by 365 Data Science

ChatGPT: How to Understand and Compete with the AI Bot

What Is A/B Testing?

Data Science for Small Business Growth (2024)

How Data Science Helps Mid-Sized Businesses in 2024

How to Build a Data-Driven Business in 2024

What are the Best Degrees in AI in 2024?

MITx Data Science Program Alternatives

Ready to start learning.

Choose your career track, build in-demand skills, and land a job in data science.

7 Key Data Science Trends For 2024-2027

You may also like:

• Important Computer Science Trends
• Top Cryptocurrency Trends
• Emerging Education Trends

Here are the 7 fastest-growing data science trends for 2024 and beyond.

We'll also outline how these trends will impact both data scientists’ work and everyday life.

Whether you’re actively involved in the data science community, or just concerned about your data privacy, these are the top trends to monitor.

1. Explosion in deepfake video and audio

Deepfakes use artificial intelligence to manipulate or create content to represent someone else.

Often this is an image or video of one person modified to someone else’s likeness.

But it can be audio too.

Back in 2019, an AI company deepfaked popular podcaster Joe Rogan’s voice so effectively it instantly went viral on social media.

And the tech has only improved since.

There’s huge scope for this technology to be used maliciously.

Another voice deep fake was used to scam a UK-based energy company out of €220,000 .

The CEO believed he was on the phone with a colleague and was told to urgently transfer the money to the bank account of a Hungarian supplier.

In fact, the call had been spoofed with deep fake technology to mimic the man’s voice and “melody”.

In fact, there's growing search interest in a practice known as "voice phishing". Which is essentially the "official" term for the practice.

As well as hoaxes and financial fraud, deepfakes can also be weaponized to discredit business figures and politicians.

Governments are starting to protect against this with legislation and social media regulation.

And with technology that can identify deepfake videos.

But the battle with deepfakes has only just begun.

2. More applications created with Python

Python is the go-to programming language for data analysis.

Why is this?

Because Python has a huge number of free data science libraries such as Pandas and machine learning libraries like Scikit-learn .

It can even be used to develop blockchain applications.

Add to this a friendly learning curve for beginners, and you have a recipe for success.

Python is now ranked as the 3rd most popular language in general by the analyst firm RedMonk.

And the popularity growth trend shows it’s on track to become number 1 within the next three years.

3. Increased demand for End-to-end AI solutions

“Dataiku” searches are up by 146% in 5 years, growing quickly even before Google acquired them.

Enterprise AI company Dataiku is now worth $4.6 billion ( according to TechCrunch ) after Google bought a stake in the company in December 2019.

The AI startup helps enterprise customers clean their large data sets and build machine learning models.

This way, companies like General Electric and Unilever can gain valuable, deep-learning insights from their massive amounts of data.

And automate important data management tasks.

Previously, businesses would have to seek expertise in all the different parts of the process and piece it together themselves.

But Dataiku handles the entire data science cycle from start to finish with a single product.

And because of this, they stand out.

Businesses want end-to-end data science solutions. And startups that provide this will eat the market.

4. Companies hire more data analysts

“Data analyst” searches are up by 265% in 5 years. Interest in this data science role displays hockey stick growth.

Demand for data analysts has shot through the roof over the last few years.

And, thanks largely to data coming in from the Internet of Things (IoT) and advances in cloud computing, global data storage is set to grow from 45 zettabytes to 175 zettabytes by 2025 .

So the need for experts to parse and analyze all of this data is set to rise.

Why are so many data analysts required?

After all, there are plenty of data analytics programs out there that can sort through it all.

And "digital transformation" has supposedly replaced many human-led business tasks.

Sure, machines can help analyze data.

But big data is often extremely messy and lacking in proper structure.

Which is why humans are needed to manually tidy training data before it is ingested by machine learning algorithms.

It’s also increasingly common for data people to be involved on the output end too.

AI-produced results are not always reliable or accurate, so machine learning companies often use humans to clean up the final data.

And write up an analysis of what they find in a way that non-tech stakeholders can understand it.

Amazon's Mechanical Turk is the biggest platform where "Turkers" complete data labeling and cleaning jobs .

The data science and machine learning methods of the 2020s will be less artificial and automated than initially expected.

Augmented intelligence and human-in-the-loop artificial intelligence will likely become a big trend in data science.

5. Data scientists joining Kaggle

Search growth for “Kaggle” has increased by 223% over 5 years. The data science platform has over 5 million users across 194 countries.

Kaggle has grown quickly to become the world's largest data science community.

And with over 8 million users across 194 countries, it’s not slowing down.

Many budding data scientists now start with Kaggle to begin their machine learning journey. 

And post the progress of their machine learning projects in real-time.

Users can even share data sets and enter competitions to solve data science challenges with neural networks.

Or work with other data scientists to build models in Kaggle’s web-based data science workbench.

Kaggle competitions can have hefty prize sums.

Academic papers have actually been published based on Kaggle competition findings too.

Successful projects from Kaggle’s hundreds of competitions will likely continue to push boundaries in the field of data science.

6. Increased interest in consumer data protection

“Data privacy”  has seen a search growth of 441% over the last 10 years. People are now searching about their data privacy in greater numbers by the month.

Consumer awareness about data privacy rose in the wake of the Cambridge Analytica scandal .

In fact, CIGI-Ipsos found that more than half of all consumers became more interested in data privacy in the year following the revelations.

Platforms like Facebook and Google, which previously harvested and shared user data freely, have since faced legal backlash and public scrutiny.

Facebook now has a large guide on privacy basics and what it does with your data.

This broader data privacy trend means that large data sets will soon be walled off and harder to come by.

Businesses and data scientists will need to navigate legislation such as the California Consumer Privacy Act which came into effect at the start of 2020.

And this could become a bane for data science when it comes to the future acquisition and use of consumer data.

7. AI devs combating adversarial machine learning

“Adversarial machine learning” searches have grown significantly in the last decade by 2,500%.

Adversarial machine learning is where an attacker inputs data into a machine learning model with the aim of causing mistakes.

Essentially, it is an optical illusion designed for a machine.

Adversarial Fashion's clothing lines trick machine-learning models with bold patterns and lettering.

Anti-surveillance clothing takes this approach to the masses.

They’re specifically designed to confuse face detection algorithms with bold shapes and patterns.

According to a Northeastern University study , this clothing can help prevent individuals' automated tracking via surveillance cameras.

Data scientists will need to defend against adversarial inputs like this. And provide trick examples for models to train on so as not to be fooled.

Adversarial training measures for models like this will become essential in the next decade.

Wrapping Up

Those are the 7 biggest data science trends over the next 3-4 years.

Data science, like any science, is changing by the day. From data governance to deepfake technology, the data science industry is set for some major shakeups.

Hopefully keeping tabs on these trends will help you stay one step ahead.

Find Thousands of Trending Topics With Our Platform

20 Data Science Topics and Areas

It is no doubt that data science topics and areas are some of the hottest business points today.

We collected some basic and advanced topics in data science to give you ideas on where to master your skills.

In today’s landscape, businesses are investing in corporate data science training to enhance their employees’ data science capabilities.

Data science topics also are hot subjects you can use as directions to prepare yourself for data science job interview questions.

1. The core of data mining process

This is an example of a wide data science topic.

What is it?

Data mining is an iterative process that involves discovering patterns in large data sets. It includes methods and techniques such as machine learning, statistics, database systems and etc.

The two main data mining objectives are to find out patterns and establish trends and relationship in a dataset in order to solve problems.

The general stages of the data mining process are: problem definition, data exploration, data preparation, modeling, evaluation, and deployment.

Core terms related to data mining are classification, predictions, association rules, data reduction, data exploration, supervised and unsupervised learning, datasets organization, sampling from datasets, building a model and etc.

2. Data visualization

Data visualization is the presentation of data in a graphical format.

It enables decision-makers of all levels to see data and analytics presented visually, so they can identify valuable patterns or trends.

Data visualization is another broad subject that covers the understanding and use of basic types of graphs (such as line graphs, bar graphs, scatter plots , histograms, box and whisker plots , heatmaps.

You cannot go without these graphs. In addition, here you need to learn about multidimensional variables with adding variables and using colors, size, shapes, animations.

Manipulation also plays a role here. You should be able to rascal, zoom, filter, aggregate data.

Using some specialized visualizations such as map charts and tree maps is a hot skill too.

3. Dimension reduction methods and techniques

Dimension Reduction process involves converting a data set with vast dimensions into a dataset with lesser dimensions ensuring that it provides similar information in short.

In other words, dimensionality reduction consists of series of techniques and methods in machine learning and statistics to decrease the number of random variables.

There are so many methods and techniques to perform dimension reduction.

The most popular of them are Missing Values, Low Variance, Decision Trees, Random Forest, High Correlation, Factor Analysis, Principal Component Analysis, Backward Feature Elimination.

4. Classification

Classification is a core data mining technique for assigning categories to a set of data.

The purpose is to support gathering accurate analysis and predictions from the data.

Classification is one of the key methods for making the analysis of a large amount of datasets effective.

Classification is one of the hottest data science topics too. A data scientist should know how to use classification algorithms to solve different business problems.

This includes knowing how to define a classification problem, explore data with univariate and bivariate visualization, extract and prepare data, build classification models, evaluate models, and etc. Linear and non-linear classifiers are some of the key terms here.

5. Simple and multiple linear regression

Linear regression models are among the basic statistical models for studying relationships between an independent variable X and Y dependent variable.

It is a mathematical modeling which allows you to make predictions and prognosis for the value of Y depending on the different values of X.

There are two main types of linear regression: simple linear regression models and multiple linear regression models.

Key points here are terms such as correlation coefficient, regression line, residual plot, linear regression equation and etc. For the beginning, see some simple linear regression examples .

6. K-nearest neighbor (k-NN) 

N-nearest-neighbor is a data classification algorithm that evaluates the likelihood a data point to be a member of one group. It depends on how near the data point is to that group.

As one of the key non-parametric method used for regression and classification, k-NN can be classified as one of the best data science topics ever.

Determining neighbors, using classification rules, choosing k are a few of the skills a data scientist should have. K-nearest neighbor is also one of the key text mining and anomaly detection algorithms .

7. Naive Bayes

Naive Bayes is a collection of classification algorithms which are based on the so-called Bayes Theorem .

Widely used in Machine Learning, Naive Bayes has some crucial applications such as spam detection and document classification.

There are different Naive Bayes variations. The most popular of them are the Multinomial Naive Bayes, Bernoulli Naive Bayes, and Binarized Multinomial Naive Bayes.

8. Classification and regression trees (CART)

When it comes to algorithms for predictive modeling machine learning, decision trees algorithms have a vital role.

The decision tree is one of the most popular predictive modeling approaches used in data mining, statistics and machine learning that builds classification or regression models in the shape of a tree (that’s why they are also known as regression and classification trees).

They work for both categorical data and continuous data.

Some terms and topics you should master in this field involve CART decision tree methodology, classification trees, regression trees, interactive dihotomiser, C4.5, C5.5, decision stump, conditional decision tree, M5, and etc.

9. Logistic regression

Logistic regression is one of the oldest data science topics and areas and as the linear regression, it studies the relationship between dependable and independent variable.

However, we use logistic regression analysis where the dependent variable is dichotomous (binary).

You will face terms such as sigmoid function, S-shaped curve, multiple logistic regression with categorical explanatory variables, multiple binary logistic regression with a combination of categorical and continuous predictors and etc.

10. Neural Networks

Neural Networks act as a total hit in the machine learning nowadays. Neural networks (also known as artificial neural networks) are systems of hardware and/or software that mimic the human brain neurons operation.

The above were some of the basic data science topics. Here is a list of more interesting and advanced topics:

11. Discriminant analysis

12. Association rules

13. Cluster analysis

14. Time series

15. Regression-based forecasting

16. Smoothing methods

17. Time stamps and financial modeling

18. Fraud detection

19. Data engineering – Hadoop, MapReduce, Pregel.

20. GIS and spatial data

For continuous learning, explore  online data science  courses for mastering these topics.

What are your favorite data science topics? Share your thoughts in the comment field above.

About The Author

Silvia Valcheva

Silvia Valcheva is a digital marketer with over a decade of experience creating content for the tech industry. She has a strong passion for writing about emerging software and technologies such as big data, AI (Artificial Intelligence), IoT (Internet of Things), process automation, etc.

Leave a Reply Cancel Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed .

• Skip to main content
• Skip to FDA Search
• Skip to in this section menu
• Skip to footer links

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

U.S. Food and Drug Administration

•   Search
•   Menu
• Science & Research
• Science and Research Special Topics
• Advancing Regulatory Science

Utilizing the UNC Clinical Data Warehouse for Assessing Efficacy, Toxicity, and Dose-Response Relationships of Molecularly Targeted Therapies in Real-World Patients

CERSI Collaborators:  Triangle CERSI, UNC-Chapel Hill: Yanguang Cao, PhD; Benyam Muluneh, PharmD

FDA Collaborators: Mirat Shah, MD; Elizabeth Duke, MD; Jonathon Vellejo, PhD; Kun Wang, PhD; Lori Ehrlich, MD, PhD; Donna Rivera PharmD, MS

CERSI Subcontractors: Flying Buttress Associates- Jeph Herrin, PhD

CERSI In-Kind Collaborators: Joshua Zeidner, MD; Rushabh Shah, PharmD, MBA

Non-Federal Entity Collaborators: Johnson and Johnson- Karla Childers, MSJ, Paul Coplan, ScD, MBA, Stephen Johnston, MSc

Project Start Date: September 1, 2023

Regulatory Science Framework:

Modernize Development & Evaluation of FDA-Regulated Products, Methods to Assess Real-World Data to serve as Real-World Evidence; Strengthen Post-market Surveillance & Labeling of FDA-Regulated Products, Methods to Assess Real-World Data to Support Regulatory Decision Making

Regulatory Science Challenge

The doses and schedules for molecularly targeted therapies are often not well-characterized prior to initiating trials with registrational intent for marketing approval. This can lead to the selection of dosages that have increased toxicity without additional efficacy. Therapies may be poorly tolerated at approved dosages, leading patients or providers to prematurely modify the dosage or discontinue an otherwise efficacious treatment. FDA launched Project Optimus to emphasize the need for dosage optimization prior to drug approval. An understanding of how the toxicities and dosage modifications for approved targeted therapies affect outcomes for patients in the real world setting will enhance Project Optimus efforts, in collaboration with the Oncology Real World Evidence program , to advance potential patient-centric uses of real-world data. To bridge this knowledge gap, researchers will investigate the prevalence of dosage modifications in patients and the resulting safety and effectiveness profiles of targeted therapies commonly prescribed for patients with Chronic Lymphocytic Leukemia (CLL) and Chronic Myeloid Leukemia (CML). They will primarily use real-world data (RWD) sourced from the Clinical Data Warehouse at the University of North Carolina at Chapel Hill to address these questions.

Project Description and Goals

In this project, researchers will address three key questions. Firstly, they will assess the frequency of dosage modifications in patients in the real-world setting using pharmacy refill records and electronic medical record information. Secondly, they will aim to determine whether modifications in treatment doses, made in response to tolerability, have any negative impact on clinical outcomes including survival. Lastly, they will explore whether the toxicity associated with treatment directly affects clinical outcomes including survival among patients in the real-world setting. The clinical data warehouse at UNC Health Care System will be primarily utilized for this project, and Triangle CERSI is also considering other sources of real-world data, such as medical claims data and randomized clinical trials, where appropriate to support the analyses. Recognizing the potential challenges and biases that can be associated with RWD studies, the capabilities of statistical modeling and advanced data mining and causal inference approaches will be harnessed to explore methodological approaches for providing valid answers to these questions.

Research Outcomes/Results

Two hundred and twenty-three patients with a mean age of 65 years completed the survey. These patients preferred a higher chance of good biopsy outcomes, and a lower chance of erectile dysfunction caused by the treatment and urinary incontinence after treatment. The patients stated in the survey that they are willing to accept:

• a 15.1%-point increase in erectile dysfunction caused by the treatment to achieve a 10%-point increase in a good biopsy outcome after HIFU ablation, and
• an 8.5%-point increase in urinary incontinence for a 10%-point increase in a good biopsy.

Also, further analysis revealed that patients who thought their cancer was more aggressive were more willing to tolerate urinary incontinence. Younger men were willing to tolerate less erectile dysfunction risk than older men. Respondents with a greater than college level of education were less willing to tolerate erectile dysfunction or urinary incontinence.

Research Impacts

Incorporating patient preference information into decisions that FDA makes about regulating devices is one of the major goals of FDA’s Center for Devices and Radiological Health (CDRH). Study findings show that patients prefer specific outcomes related to prostate ablation therapies like HIFU. The study results may help inform the design and regulation of current and future prostate tissue ablation devices by providing information about outcomes that patients most desire.

Publications

• PMID: 34677594; Citation: Wallach JD, Deng Y, McCoy RG, Dhruva SS, Herrin J, Berkowitz A, Polley EC, Quinto K, Gandotra C, Crown W, Noseworthy P, Yao X, Shah ND, Ross JS, Lyon TD. Real-world Cardiovascular Outcomes Associated With Degarelix vs Leuprolide for Prostate Cancer Treatment.  JAMA Netw Open. 2021;4(10):e2130587. doi:10.1001/jamanetworkopen.2021.30587 .
• PMID: 36191949; Citation: Deng Y, Polley EC, Wallach JD, Dhruva SS, Herrin J, Quinto K, Gandotra C, Crown W, Noseworthy P, Yao X, Lyon TD, Shah ND, Ross JS, McCoy RG. Emulating the GRADE trial using real world data: retrospective comparative effectiveness study. BMJ . 2022 Oct 3;379:e070717. doi: 10.1136/bmj-2022-070717 .

Read our research on: Gun Policy | International Conflict | Election 2024

Regions & Countries

Political typology quiz.

Notice: Beginning April 18th community groups will be temporarily unavailable for extended maintenance. Thank you for your understanding and cooperation.

Where do you fit in the political typology?

Are you a faith and flag conservative progressive left or somewhere in between.

Take our quiz to find out which one of our nine political typology groups is your best match, compared with a nationally representative survey of more than 10,000 U.S. adults by Pew Research Center. You may find some of these questions are difficult to answer. That’s OK. In those cases, pick the answer that comes closest to your view, even if it isn’t exactly right.

About Pew Research Center Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of The Pew Charitable Trusts .