Programme

P.21 -- Poster, ANSWCCMS-2023

Date: Tuesday, 7 November 2023

Time: 13:00 - 14:30

Studying And Predicting Energy Gap Of Materials By Machine Learning Method

Le Huu Nghia

College of Natural Sciences - Can Tho University

Today, big data and artificial intelligence are revolutionizing many areas of our lives and the sciences. Materials science is not an exception, data-driven materials science is becoming the fourth paradigm of materials research, this report uses the Machine Learning (ML) method to find the best high-performance energy band gap in materials data. In this work, we start with data processing, analysis, modeling, and visualization of the results as well as a data organization process for ML models in materials research. We change the percentage of the dataset by setting a random seed from 0 to 60 (random. seed()) and find the best percentage for the data with train split = 70%; test split = 10%, and validation split = 20%. From the above percentage, our supervised ML finds the best-performing model: ExtraTreesRegressor() with the parameters: Coefficient of determination (R2): 0.7479, Mean Absolute Error (MAE): 0.5844, Root Mean Squared Error (RMSE): 0.8088. We consider therefore the different facets of interpretability prediction of models of ML and their importance in materials science. Finally, we propose solutions and future research paths for various challenges in computational materials science.

Presenter: Le Huu Nghia