Fig. 3

Accuracy of machine learning classifiers. Prediction accuracies (y-axis) of the various predictors as a function of the feature set used for training (x-axis). The training sets consist of structural properties (S1, S2, S3, S4, S4, S6), energetics (E1, E2) and a combination of both (C). Refer to Table 1 for the detailed list of features included in each set. Five different machine learning algorithms have been used for the training: Bagging, Random Forest, Adaptive Boosting, Gradient Boosting and Neural Network, reported in blue, purple, green, red and brown, respectively