|About this Abstract
||2019 TMS Annual Meeting & Exhibition
||Computational Approaches for Big Data, Artificial Intelligence and Uncertainty Quantification in Computational Materials Science
||Materials Science Learning and Discovery from Large-scale Text Mining
||Leigh Weston, Vahe Tshitoyan, John Dagdelen, Kristin A. Persson, Gerbrand Ceder, Anubhav Jain
|On-Site Speaker (Planned)
The wealth of human materials science knowledge is currently stored in millions of published scientific articles. The magnitude is so large that individual materials scientists will only access a fraction of published data in their lifetime. Recent developments in Artificial Intelligence and Natural Language Processing (NLP) have enabled the development of tools capable of extracting and compressing this information. We have used NLP, as well as supervised and unsupervised machine learning, to extract the key information from millions of scientific abstracts, including: materials properties, crystallographic phase, device applications, and other important characteristics. We will show that the information extracted from these texts can be used to compress all materials science knowledge into a database that is more readily accessible to researchers, and for large scale data mining for predicting new materials, as well as their properties and applications.
||Planned: Supplemental Proceedings volume