Machine learning and large language models extract materials data from science texts
Automated Materials Data and Knowledge Extraction using Machine Learning and Large Language Models
Artificial Intelligence
Summary
Extracting useful data about materials from scientific papers is hard and time-consuming. The authors review how machine learning and language models have evolved to automatically pull information like materials properties from text, tables, and images. These advances help build big searchable databases and enable chatbots that can answer questions about materials. The paper also discusses challenges like inconsistent terminology and the lack of standard test datasets. It offers practical advice for making more reliable extraction tools.
What this means in practice
- •For materials data engineers: Build scalable pipelines that reliably extract materials information from scientific papers for large databases and knowledge graphs.
- •For laboratory automation teams: Integrate AI-driven data extraction with autonomous labs to support closed-loop materials discovery and optimization workflows.
- •For chemical informatics software developers: Develop research assistant chatbots that answer generic materials science questions using multimodal scientific literature inputs.$Commercial implications: Enables AI chatbots offering materials research support services to institutions and companies, improving research efficiency.
A survey. It maps existing work.
Authors
Aravindan Kamatchi Sundaram, B. Pabitramohan Prusty, Rohit Batra
Abstract
Over the past decade, substantial efforts have been devoted to developing machine learning (ML) methods for extracting materials data and, where possible, materials knowledge directly from the scientific literature. This review traces the evolution of these methods from heuristic, rule-based pipelines through natural language processing (NLP) models and transformer-based encoders to, lastly, large language models (LLMs) and agentic systems capable of extracting materials information across multiple modalities, including text, tables, and figures. Beside construction of large, queryable materials databases and knowledge graphs that support downstream materials discovery, we discuss how these advances have enabled new capabilities, such as, generic question answering across materials science, AI co-scientists and chatbots for research assistance, and closed-loop optimization and discovery when integrated with autonomous laboratories. We critically assess the datasets generated using these approaches, their extraction accuracy and computational costs, and the respective strengths and limitations of different methodologies, and even provide practical guidelines for developing more reliable and scalable materials data extraction pipelines. Finally, we identify key open challenges, including the extraction of full materials structure-property-processing-performance tetrahedron at the sample level, the scarcity of reliable benchmark datasets for systematic evaluation, and the resolution of non-standard materials terminology across different document modalities.