Materials Data Extraction from Literature by NLP and Large Language Models

View at publisher Summarise with VIRAT
Authors: Xue Jiang, Yanjing Su
Cited by: 0
References: 46

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This chapter explores the transformative impact of natural language processing (NLP) and large language models (LLMs) on extracting structured data from scientific literature in materials research. The increasing volume of publications necessitates automated methods, as manual extraction is inefficient. The chapter reviews traditional NLP pipelines, including corpus acquisition, text preprocessing, and named entity recognition (NER), and contrasts these with emerging LLM-based approaches. Case studies of tools like ChemDataExtractor and MaterialsBERT illustrate practical applications in extracting data on chemical compounds, polymers, and alloys. The standardized workflow of NLP in scientific literature is discussed, emphasizing its complexity due to domain-specific terminology. The chapter concludes by comparing conventional NLP and LLM-based methods, underscoring the latter's strengths in semantic understanding and rapid adaptation across diverse material domains. The potential of integrating classical NLP with LLMs to advance data-driven research in materials science is highlighted. The chapter explores automated information extraction from scientific literature for materials science, focusing on conventional and LLM-based NLP pipelines. It begins with corpus collection and preprocessing, including cleaning, tokenization, and part-of-speech tagging, to prepare data for extraction.

Fields of Study & Subjects

Machine Learning in Materials Science Topic Modeling Biomedical Text Mining and Ontologies

Publication details

Journal / source
Wiley eBooks
Pages
205–218
Publication date
17 March 2026
Language
EN
Document type
Book chapter