Material Named Entity Recognition (MNER) for Knowledge-driven Materials Using Deep Learning Approach

TL;DR

This paper presents a deep learning model based on Bi-LSTM for Material Named Entity Recognition (MNER) to extract valuable knowledge from scientific literature in materials science, achieving high accuracy and aiding data-driven discovery.

Contribution

It introduces a Bi-LSTM based neural network model specifically designed for MNER in materials science literature, demonstrating high performance and providing insights for future research.

Findings

Achieved approximately 97% F1 score on MNER task

Demonstrated effectiveness of Bi-LSTM in extracting material entities

Provided detailed analysis and future research directions

Abstract

The scientific literature contains a wealth of cutting-edge knowledge in the field of materials science, as well as useful data (e.g., numerical data from experimental results, material properties and structure). These data are critical for data-driven machine learning (ML) and deep learning (DL) methods to accelerate material discovery. Due to the large and growing number of publications, it is difficult for humans to manually retrieve and retain this knowledge. In this context, we investigate a deep neural network model based on Bi-LSTM to retrieve knowledge from published scientific articles. The proposed deep neural network-based model achieves an f-1 score of \~97\% for the Material Named Entity Recognition (MNER) task. The study addresses motivation, relevant work, methodology, hyperparameters, and overall performance evaluation. The analysis provides insight into the results of the experiment and points to future directions for current research.