Fundamental band gap and alignment of two-dimensional semiconductors explored by machine learning
Zhu Zhen1, †, Dong Baojuan2, 4, 5, Guo Huaihong3, Yang Teng2, ‡, Zhang Zhidong2
       

Comparison of calculated HSE values with predicted fundamental band gaps and band edges by LR, RFR, and SVR models. These models are evaluated with 10-fold cross-validation and only the predicted results of the validation sets are shown. Each column of the subfigures represents one predictive model and each row shows one selected electronic property. Selected signatures of constituent elements are used as predictors (predictors set-II) for these machine learning models. The dashed lines are guide to the eyes, representing that the predicted values are equal to the computed HSE data.