Table A7.

Ensemble model’s performance for the five-fold cross-validation.

	Target Class	Precision	Recall	F1-Score	AUC Score	Taken Weight	Accuracy		Confusion Matrix
							Train	Test
Fold 1	Class 0	0.93	0.91	0.92	-	4 (Xgb) 3 (RF)	-	-	True label	Predicted label
	Class 1	0.84	0.87	0.85	-		-	-		91	9
	Average	0.88	0.89	0.89	0.94		.99	0.90		7	47
Fold 2	Class 0	0.93	0.87	0.90	-	1 2	-	-	True label	Predicted label
	Class 1	0.78	0.87	0.82	-		-	-		87	13
	Average	0.85	0.87	0.86	0.94		1.00	0.87		7	47
Fold 3	Class 0	0.91	0.91	0.91	-	1 1	-	-	True label	Predicted label
	Class 1	0.83	0.83	0.83	-		-	-		91	9
	Average	0.87	0.87	0.87	0.95		1.00	0.88		9	45
Fold 4	Class 0	0.96	0.91	0.93	-	1 4	-	-	True label	Predicted label
	Class 1	0.84	0.92	0.88	-		-	-		91	9
	Average	0.90	0.92	0.91	0.96		1.00	0.91		4	49
Fold 5	Class 0	0.95	0.94	0.94	-	2 2	-	-	True label	Predicted label
	Class 1	0.89	0.91	0.90	-		-	-		94	6
	Average	0.92	0.92	0.92	0.96		1.00	0.92		5	48
	All folds’ average	0.88	0.89	0.89	0.95		0.99	0.90