Auto-Generating Databases about Thermoelectric Materials and Their Properties: Towards Data-Driven Materials Discovery
DescriptionThis talk will introduce the ‘chemistry aware’ natural-language-processing software tool, ChemDataExtractor,[1,2] and illustrate its ability to auto-generate a database on thermoelectric materials and their properties. The database contains a total of 22,805 records of thermoelectric data that have been mined from the academic literature. Specifically, it includes chemical names of thermoelectric materials as well as five of their cognate properties: ZT; thermal conductivity, κ; Seebeck coefficient, S; electrical conductivity, σ; power factor, PF. The ability to use this database to forecast research trends about thermoelectric materials is demonstrated. The underpinning database auto-generation methods are also discussed in terms of their ability to interface with machine-learning pipelines which enable a design-to-device approach to data-driven materials discovery.
References to our AI tools
 Swain and Cole, J. Chem. Inf. Model. 2016, 56, 10, 1894–1904 www.chemdataextractor.org
 Mavracic, Court, Isazawa, Elliott, Cole, J. Chem. Inf. Model. 2021, 61, 9, 4280–4289 www.chemdataextractor2.org
 Sierepeklis, Cole, Sci. Data (Springer Nature), 2022, 9, 248
 Cole, Acc. Chem. Res. 2020, 53, 599-610.
TimeMonday, June 2616:30 - 17:00 CEST