Abstract This teaching case explores the application of advanced machine learning algorithms,XGBoost and LightGBM,in the high-throughput screening of perovskite materials for thermal stability and bandgap,aiming to integrate the intersection of computational chemistry and data science into chemical education.Through practical case analysis,this article not only deepens students’ understanding of the basic theories of computational chemistry but also enhances their ability to apply machine learning technologies to solve chemical problems,promoting interdisciplinary thinking and skill cultivation.Additionally,SHAP analysis is utilized to enhance the interpretability of model predictions,providing students with valuable opportunities to understand and master cutting-edge technologies,and laying a solid foundation for their research and professional development.
ZHANG Zhao-Sheng. Innovative Application of Machine Learning in Computational Chemistry Education: High-Throughput Screening Study on Perovskite Materials[J]. Chinese Journal of Chemical Education, 2024, 45(22): 97-103.
