ChemNLP: A Natural Language-Processing-Based Library for Materials Chemistry Text Data

Abstract

In this work, we present the ChemNLP library that can be used for ( 1) curating open access datasets for materials and chemistry literature, developing and comparing traditional machine learning, transformers and graph neural network models for (2) classifying and clustering texts, (3) named entity recognition for largescale text-mining, (4) abstractive summarization for generating titles of articles from abstracts, (5) text generation for suggesting abstracts from titles, (6) integration with density functional theory dataset for identifying potential candidate materials such as superconductors, and (7) web-interface development for text and reference query. We primarily use the publicly available arXiv and PubChem datasets, but the tools can be used for other datasets as well. Moreover, as new models are developed, they can be easily integrated in the library.

Flexible, model-agnostic method for materials data extraction from text using general purpose language models

“…So far, the majority of materials data extraction approaches focus on fully automatic data extraction. 2–7 Automation is clearly desirable, particularly when extracting very large databases. However, more automation tends to require more complexity in the software, sophistication in training schemes, and knowledge about the extracted property.…”

Accelerating Dental Adhesive Innovations Through Active Learning and Bayesian Optimization

“…Of late, applications of machine learning to materials science and engineering has emerged as a transformative tool for materials discovery [1,2,3]. These applications typically range from deep learning [4,5], for image-based tasks [6,7,8], to more recent natural language processing adaptations to aid in design of new materials [9,10,11]. Although machine learning approaches to materials discovery have found applications across multiple materials domains, it has yet to make an impact in the field of dental materials [12,13].…”

InterMat: accelerating band offset prediction in semiconductor interfaces with DFT and deep learning

“…50,51 Such models have often been applied for bulk property predictions and their applicability for defects and interfaces remains an open question. Several machine learning tools available in JARVIS such as classical force-field inspired descriptors (CFID), 43 atomistic line graph neural network (ALIGNN), 52,53 computer vision for atomistic images (AtomVision) 54 and natural language processing for chemistry (ChemNLP) 55,56 can be used in this regards to accelerate the interface design tasks. In particular, ALIGNN has been used to develop several fast surrogate models for property predictions as well as a unified force-field for fast structure optimizations.…”