表 7.

本研究所提出的方法与八种集成分类器结果比较

Comparison of the results of the proposed method and 8 ensemble classifiers

Classifier	Feature selection methods	AUC	ACC	SEN	SPE
AdaBoost	MRMR	0.659	0.621	0.652	0.591
Bagging	MIFS	0.731	0.666	0.668	0.667
CatBoost	MIFS	0.679	0.675	0.808	0.535
Extra Trees	MRMR	0.704	0.693	0.753	0.631
GBDT	t_score	0.676	0.649	0.756	0.538
LightGBM	t_score	0.652	0.639	0.736	0.538
Random Forest	MIFS	0.710	0.675	0.721	0.631
XGBoost	CMIM	0.620	0.602	0.615	0.593
The proposed method		0.828	0.766	0.877	0.648