MORE | Spring 2023
Thermoelectric Material Discovery Using Machine Learning
The conventional approach of discovering thermoelectric materials from an essentially infinite universe of chemical compounds includes costly and time-consuming studies. In the current study, thermoelectric properties (labels) and descriptors (features) of a group of substances are calculated using density functional theory (DFT) simulations. With the knowledge that machine learning (ML) can extract the key characteristics from the small database and we attempt to use them to predict the behavior of any new compound.
Hometown: Tempe, Arizona, United States
Graduation date: Spring 2023