TABLE III.

Optimized Hyperparameters for Each Classifier

Classifier	Hyperparameter description	Optimized hyperparameters
LR	C: inverse of regularization strength	C=0.5
SVM	Kernel: type of decision function C: penalty of the error term γ: parameter of radial basis kernel function	Kernel: radial basis function C=50 γ= 0.001
RF	Nleaf_min: minimum number of samples required to he at a lead node Nfeat_max: max number of features allowed to form each tree	Nleaf_min = 1 Nfeat_max 0.2·N
kNN	Nn: number of neighbors	Nn= 5

LR = Logistic Regression; SVM = Support vector machine; RF = Random Forest; kNN = k Nearest Neighbour; N=total number of features; N= total number of features