Case Studies of Alloy Information Extraction

View at publisher Summarise with VIRAT
Authors: Xue Jiang, Yanjing Su
Cited by: 0
References: 78

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 discusses the development and application of automated text mining pipelines for extracting structured data from scientific literature on alloys. The rapid growth of publications in materials science has made traditional manual data curation inefficient. NLP and ML techniques are employed to systematically capture chemical compositions, processing routes, and material properties from thousands of articles. The chapter highlights rule-based and semi-supervised approaches for named entity recognition (NER), relation extraction, and table parsing, which enable high-accuracy construction of databases without requiring large, annotated corpora. By integrating these methods, the resulting datasets support fundamental understanding of composition-property relationships and facilitate predictive modeling for alloy design. The chapter also presents a case study on superalloys, demonstrating the potential of automated pipelines to accelerate data-driven materials research. The automated text mining pipeline for superalloys involves multiple stages: document download, preprocessing, table parsing, text classification, named entity recognition (NER), relation extraction, and interdependency resolution. The process begins with scientific articles in XML, HTML, or plain-text formats, which are preprocessed to assemble complete document records and remove irrelevant content.

Fields of Study & Subjects

Machine Learning in Materials Science Topic Modeling Advanced Text Analysis Techniques

Publication details

Journal / source
Wiley eBooks
Pages
243–298
Publication date
17 March 2026
Language
EN
Document type
Book chapter