Table 4.

Performance comparison of the DXLR model with removal of network feature separately

Method	Metric [Mean ± SD]
	Precision	Recall	Accuracy	F1 score	AUC
DXLR	0.723 ± 0.014	0.892 ± 0.012	0.857 ± 0.007	0.798 ± 0.010	0.934 ± 0.004
−Node	0.699 ± 0.014	0.905 ± 0.010	0.846 ± 0.006	0.788 ± 0.009	0.927 ± 0.004
−Edge	0.681 ± 0.013	0.907 ± 0.012	0.836 ± 0.006	0.778 ± 0.009	0.923 ± 0.004
−Rank	0.492 ± 0.015	0.663 ± 0.019	0.676 ± 0.010	0.565 ± 0.012	0.737 ± 0.010