TABLE 2.

The evaluation criteria for 14 machine learning models.

Model	Accuracy	AUC a	Recall	Prec b	F1 c	Kappa d	TT e
Light Gradient Boosting Machine	0.9895	0.9998	0.9462	0.9877	0.9878	0.9842	68.718
Logistic Regression	0.991	0.9984	0.9474	0.9891	0.9893	0.9865	3.019
Random Forest Classifier	0.9895	0.9981	0.9349	0.9848	0.9866	0.9842	0.917
Extra Trees Classifier	0.991	0.9976	0.9474	0.9891	0.9893	0.9865	1.065
Gradient Boosting Classifier	0.9789	0.996	0.9268	0.9782	0.9775	0.9685	153.012
K Neighbors Classifier	0.9864	0.9959	0.967	0.9895	0.9871	0.9798	1.086
Naive Bayes	0.9759	0.9814	0.8429	0.967	0.9705	0.9638	0.887
Decision Tree Classifier	0.9698	0.9775	0.8718	0.965	0.9661	0.9547	1.33
Ada Boost Classifier	0.6779	0.966	0.6942	0.7457	0.661	0.5924	8.818
Linear Discriminant Analysis	0.6029	0.7268	0.603	0.6543	0.609	0.4357	1.593
Quadratic Discriminant Analysis	0.565	0.7017	0.4901	0.3913	0.4428	0.3945	1.17
Dummy Classifier	0.4003	0.5	0.25	0.1603	0.2289	0	0.766
SVM - Linear Kernel	0.991	0	0.9474	0.9891	0.9893	0.9865	0.878
Ridge Classifier	0.991	0	0.9474	0.9891	0.9893	0.9865	0.872

^a AUC: area under curve, the area under the ROC, curve.

^b Prec.: precision.

^c F1: F1-score, the harmonic mean of precision and recall value.

^d Kappa: Kappa-value, measuring model evaluation accuracy in multiple classifications.

^e TT: the processing time for the 10-fold training.