Table 1.

The performance of the best model for each ML method obtained from different feature encodings.

Methods	Features	MCC	Accuracy	Sensitivity	Specificity	AUC
ERT	H8 (420)	0.546	0.772	0.740	0.805	0.813
RF	H5 (420)	0.546	0.776	0.829	0.724	0.805
GB	H10 (577)	0.545	0.772	0.789	0.756	0.806
AB	H5 (420)	0.531	0.764	0.715	0.813	0.767
SVM	H4 (597)	0.457	0.728	0.772	0.683	0.746

The first column represents the method name developed in this study. The second column represents the hybrid model and its corresponding number of features. The third, fourth, fifith, sixth, and seventh columns, respectively, represent the MCC, accuracy, sensitivity, specificity, and AUC. RF: random forest; ERT: extra tree classifier; SVM: support vector machine; GB: gradient boosting; and AB: adaBoost.