TABLE 5.

Performance values of proposed models calculated on the testing dataset.

Trained model for feature selection	Classification model	Accuracy	Std of accuracy	Precision	Std of precision	Recall	Std of recall	Specificity	Std of specificity
RF	RF	86.02%	10.93	90.25%	6.33	85.77%	11.25	72.44%	23.44
	LR	85.07%	8.95	86.98%	7.82	84.91%	9.1	80.56%	18.15
	GB	83.52%	8.11	85.30%	7.09	83.37%	8.34	77.56%	18.81
	SVM	85.00%	10	89.15%	6.04	84.78%	10.13	70.56%	20.51
LR	RF	83.02%	10.8	85.93%	7.98	82.81%	11.12	71.44%	22.43
	LR	82.52%	8.77	83.78%	8.53	82.32%	8.89	77.56%	14.62
	GB	80.52%	9.9	83.19%	9.56	80.36%	10.23	75.44%	19.38
	SVM	84.02%	8.63	88.06%	5.49	83.77%	8.99	70.44%	19.37
GB	RF	87.50%	12.09	90.22%	9.69	87.22%	12.41	75.44%	22.71
	LR	83.07%	11.24	84.19%	10.89	82.96%	11.4	81.56%	18.9
	GB	83.55%	9.27	85.06%	8.61	83.31%	9.59	79.44%	19.39
	SVM	84.50%	10.11	88.86%	5.94	84.22%	10.4	69.44%	21.1
SVM	RF	84.02%	12.02	87.22%	9.04	83.77%	12.28	73.44%	24.36
	LR	88.52%	6.75	89.50%	6.79	88.38%	6.88	83.67%	10.69
	GB	80.60%	8.5	82.88%	7.44	80.45%	8.82	76.44%	19.84
	SVM	84.50%	10.59	87.01%	8.57	84.27%	10.93	72.44%	19.73