Superconducting materials are widely used in industry and academia because they are uniquely capable of conducting electric current with zero resistance at or below a certain critical temperature T<sub>c</sub>. We demonstrated that one of the most recent studies (“A data-driven statistical model for predicting the critical temperature of a superconductor” by K. Hamidieh, Computational Materials Science, 2018, 154, 346) reports on models that were based on the dataset that contains 27% of duplicate entries. In this study, we have developed a series of (quantitative)structure-property relationship models for data on critical temperatures and types of high- and low-temperature superconductive inorganic materials. Both traditional machine learning techniques (random forests, neural networks, generalized linear model, multiple linear regression, gradient boosting decision trees, and support vector machine) and CORAL modeling approach were used to generate predictive models.

Predictive modeling of critical temperatures in superconducting materials

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.