A Data-Driven Approach for Predicting the Elastic Properties of Inorganic Materials

The goal of this project is to aid in the rapid design and screening of new materials by bypassing the need for first-principle computational methods and, instead, use fast supervised machine learning algorithms to predict the elastic modulus of inorganic materials. We show that simple machine learning algorithms can be used to predict elastic moduli with relatively high accuracy, achieving a coefficient of correlation of 0.9 and low RMSE.