A rule-free workflow for the automated generation of databases from scientific literature

Abstract

In recent times, transformer networks have achieved state-of-the-art performance in a wide range of natural language processing tasks. Here we present a workflow based on the fine-tuning of BERT models for different downstream tasks, which results in the automated extraction of structured information from unstructured natural language in scientific literature. Contrary to existing methods for the automated extraction of structured compound-property relations from similar sources, our workflow does not rely on the definition of intricate grammar rules. Hence, it can be adapted to a new task without requiring extensive implementation efforts and knowledge. We test our data-extraction workflow by automatically generating a database for Curie temperatures and one for band gaps. These are then compared with manually-curated datasets and with those obtained with a state-of-the-art rule-based method. Furthermore, in order to showcase the practical utility of the automatically extracted data in a material-design workflow, we employ them to construct machine-learning models to predict Curie temperatures and band gaps. In general we find that, although more noisy, automatically extracted datasets can grow fast in volume and that such volume partially compensates for the inaccuracy in downstream tasks.

arXiv comment19 pages, 11 figures

References in corpus (13)

• Electric Field Effect in Atomically Thin Carbon Films
• BioBERT: a pre-trained biomedical language representation model for biomedical text mining
• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
• Big Data of Materials Science - Critical Role of the Descriptor
• Universal Fragment Descriptors for Predicting Electronic Properties of Inorganic Crystals
• Band structure of MoS_2, MoSe_2, and alpha-MoTe_2: Angle-resolved photoelectron spectroscopy and ab-initio calculations
• Materials Cloud, a platform for open computational science
• Benchmarking Materials Property Prediction Methods: The Matbench Test Set and Automatminer Reference Algorithm
• Materials Cartography: Representing and Mining Material Space Using Structural and Electronic Fingerprints
• Learning Properties of Ordered and Disordered Materials from Multi-fidelity Data
• A general-purpose material property data extraction pipeline from large polymer corpora using Natural Language Processing
• Predicting the Curie temperature of ferromagnets using machine learning

Cited by in corpus (4)

• Extracting Accurate Materials Data from Research Papers with Conversational Language Models and Prompt Engineering
• The Northeast Materials Database for Magnetic Materials
• Towards an automated workflow in materials science for combining multi-modal simulative and experimental information using data mining and large language models
• A New Workflow for Materials Discovery Bridging the Gap Between Experimental Databases and Graph Neural Networks