Table 3.

Results based on feature selection using random forest algorithm.

Features	Algorithm	Accuracy	Sensitivity	Specificity
5	Logistic	0.6891	0.813	0.5187
	k-NN	0.6687	0.7565	0.5502
	Tree	0.644	0.6866	0.5902
	R-forest	0.7139	0.7645	0.6471
	SVM	0.7143	0.7877	0.6156
	NN	0.7198	0.7522	0.5869
10	Logistic	0.6985	0.7752	0.5956
	k-NN	0.6867	0.7284	0.6315
	Tree	0.652	0.7053	0.5813
	R-forest	0.7122	0.6992	0.7347
	SVM	0.6997	0.7729	0.6024
	NN	0.6913	0.6677	0.5889
15	Logistic	0.7118	0.7582	0.6504
	k-NN	0.6952	0.7125	0.6745
	Tree	0.6505	0.7029	0.5812
	R-forest	0.7418	0.7267	0.7669
	SVM	0.7333	0.7867	0.662
	NN	0.701	0.6759	0.5916
20	Logistic	0.6934	0.7475	0.6212
	k-NN	0.6756	0.7124	0.6289
	Tree	0.6478	0.7087	0.5662
	R-forest	0.7427	0.7364	0.7549
	SVM	0.7314	0.7932	0.6483
	NN	0.6999	0.6772	0.5867
25	Logistic	0.6864	0.7457	0.605
	k-NN	0.6874	0.7263	0.6364
	Tree	0.647	0.7005	0.5731
	R-forest	0.7441	0.7467	0.7446
	SVM	0.7282	0.7934	0.6391
	NN	0.6911	0.6766	0.5905
30	Logistic	0.6914	0.7272	0.6441
	k-NN	0.69	0.7294	0.6386
	Tree	0.6472	0.7046	0.5682
	R-forest	0.7447	0.7431	0.7512
	SVM	0.7234	0.7899	0.6344
	NN	0.7024	0.6907	0.5878