Table 7.

Performance comparison of different model combinations

Type	Methods	Error rate	AUC	F1
BBLC-based	Logistic Regression	0.3563	0.7005	0.7465
	Support Vector Machine	0.2706	0.7292	0.7568
	Random Forest	0.2419	0.7534	0.7815
	Customized XGBoost	0.2131	0.7992	0.8075
HDL-based	Logistic Regression	0.2546	0.7378	0.7863
	Support Vector Machine	0.2234	0.7681	0.8077
	Random Forest	0.1623	0.8357	0.8548
	Customized XGBoost	0.1012	0.8639	0.8927