Table 4.

The classification performance (in percentage) of four machine learning methods with GM and WM.

			GM + WM
	RetainedFeas	OptimalFeas	Sen	Spec	Acc	AUC
SPEC	10,740	4,115	89.20 ± 2.31	79.26 ± 4.66	84.20 ± 3.01	84.23 ± 3.74
ReliefF	10,424	3,996	89.66 ± 3.84	80.01 ± 3.10	84.92 ± 1.80	84.84 ± 3.48
RFE	11,372	5,647	86.29 ± 3.07	78.47 ± 4.30	82.37 ± 3.26	82.38 ± 3.96
STABLASSO	4,425	2,753	87.43 ± 2.50	80.67 ± 1.72	84.25 ± 1.43	84.05 ± 1.88

The best performance for each indicator is shown in bold.