TABLE 7.

Comparison of performance of different combinations of stacking ensemble models.

Classifier	Accuracy	F1-score	Precision	Recall	Time(s)	AUC-ROC	TPR
NB	0.74	0.73	0.74	0.73	1	0.81	0.73
CatBoost	0.88	0.88	0.88	0.88	122	0.96	0.88
DT	0.88	0.88	0.87	0.89	7	0.88	0.89
GB	0.84	0.84	0.86	0.8	83	0.92	0.86
NDCG	0.87	0.87	0.87	0.87	455	0.97	0.89
GNCD	0.83	0.83	0.84	0.82	1,005	0.83	0.82
Hard voting	0.87	0.86	0.88	0.85	182	0.87	0.85
Soft voting	0.88	0.88	0.86	0.91	187	0.88	0.91