TABLE 1.

Performance for four different classifiers with optimized hyperparameters.

Dataset	Classifiers	AUC	MCC	F1	Cohen’s kappa	Balanced Accuracy	Accuracy	Sensitivity	Specificity
Training	DNN	0.802	0.497	0.809	0.493	0.741	0.761	0.851	0.630
	KNN	0.762	0.441	0.789	0.436	0.713	0.735	0.834	0.591
	SVM	0.778	0.478	0.805	0.472	0.729	0.753	0.856	0.602
	RF	0.771	0.549	0.837	0.491	0.727	0.774	0.977	0.476
IV	DNN	0.798	0.458	0.795	0.453	0.721	0.743	0.839	0.603
	KNN	0.764	0.409	0.778	0.405	0.698	0.721	0.821	0.574
	SVM	0.777	0.455	0.804	0.438	0.709	0.743	0.888	0.529
	RF	0.747	0.502	0.824	0.436	0.700	0.752	0.975	0.424

The best performance values among the classifiers.