Table 3.

Comparison of mean AUC of combination of four features and ten classifiers.

Classifier	Feature
	SE	FE	AE	PE	Mean ± SD
AB	0.808 ± 0.044	0.904 ± 0.027	0.804 ± 0.053	0.720 ± 0.080	0.809 ± 0.085
DT	0.886 ± 0.033	0.946 ± 0.025	0.883 ± 0.038	0.817 ± 0.060	0.883 ± 0.061
GP	0.743 ± 0.059	0.865 ± 0.037	0.736 ± 0.069	0.667 ± 0.077	0.753 ± 0.095
LS	0.690 ± 0.053	0.825 ± 0.055	0.674 ± 0.068	0.584 ± 0.098	0.693 ± 0.111
GNB	0.726 ± 0.063	0.857 ± 0.036	0.720 ± 0.073	0.609 ± 0.090	0.728 ± 0.111
KNN	0.847 ± 0.044	0.921 ± 0.025	0.847 ± 0.050	0.775 ± 0.099	0.848 ± 0.080
MLP	0.716 ± 0.063	0.850 ± 0.040	0.709 ± 0.075	0.615 ± 0.092	0.722 ± 0.109
QDA	0.726 ± 0.063	0.857 ± 0.036	0.720 ± 0.073	0.622 ± 0.090	0.731 ± 0.108
RF	0.936 ± 0.031	0.969 ± 0.021	0.937 ± 0.031	0.874 ± 0.111	0.929 ± 0.070
RS	0.0728 ± 0.062	0.859 ± 0.036	0.721 ± 0.074	0.610 ± 0.087	0.729 ± 0.111
Mean ± SD	0.780 ± 0.095	0.885 ± 0.057	0.775 ± 0.104	0.689 ± 0.132

Boldface indicates FE + RF is the optimal method.