Table 5.

Comparison of learning algorithms on reduced one-hot encoded features

Models	12 one-hot features					7 one-hot features
	Accuracy	Precision	Recall	Specificity	F1 score	Accuracy	Precision	Recall	Specificity	F1-score
DT	0.761	0.83	0.72	0.81	0.77	0.775	0.82	0.78	0.77	0.79
RF	0.873	0.94	0.82	0.94	0.88	0.873∗	0.9	0.88	0.87	0.89
KNN	0.929∗	0.95∗	0.93∗	0.94∗	0.94∗	0.929	0.97	0.9	0.97∗	0.94
Naïve Bayesian	0.915	0.95	0.9	0.94	0.92	0.915	0.97	0.88	0.97∗	0.92
SVM	0.929∗	0.95∗	0.93∗	0.94∗	0.94∗	0.944∗	0.95∗	0.95∗	0.94	0.95∗
ANN	0.915	0.95	0.9	0.94	0.92	0.901	0.95	0.88	0.94	0.91
Average	0.887	0.93	0.87	0.92	0.90	0.890	0.93	0.88	0.91	0.90

ANN = artificial neural network; DT = decision tree; KNN = k-nearest neighbor; RF = random forest; SVM = support vector machine.

^∗Indicates the highest accuracy value for each measurement.