Table 4.

Results based on feature selection using logistic regression algorithm.

Features	Algorithm	Accuracy	Sensitivity	Specificity
5	Logistic	0.6914	0.7792	0.5707
	k-NN	0.6606	0.7004	0.6102
	Tree	0.6357	0.698	0.5512
	R-forest	0.6675	0.6587	0.6862
	SVM	0.6757	0.7182	0.622
	NN	0.6826	0.688	0.59
10	Logistic	0.6926	0.7449	0.6212
	k-NN	0.6777	0.7171	0.6275
	Tree	0.6219	0.7025	0.5122
	R-forest	0.6903	0.6888	0.6967
	SVM	0.7005	0.7706	0.605
	NN	0.6777	0.6695	0.5902
15	Logistic	0.6875	0.7269	0.6351
	k-NN	0.6557	0.6811	0.6252
	Tree	0.6205	0.7353	0.4599
	R-forest	0.7009	0.7022	0.7036
	SVM	0.7024	0.7829	0.5929
	NN	0.6846	0.666	0.5893
20	Logistic	0.6758	0.7113	0.6288
	k-NN	0.6605	0.7017	0.6082
	Tree	0.6151	0.7632	0.4104
	R-forest	0.7021	0.7281	0.6698
	SVM	0.6973	0.8035	0.5531
	NN	0.6695	0.6666	0.5876
25	Logistic	0.6552	0.6742	0.6308
	k-NN	0.6403	0.7053	0.5545
	Tree	0.61	0.7525	0.4132
	R-forest	0.7125	0.7427	0.6735
	SVM	0.6952	0.8041	0.5463
	NN	0.6721	0.6562	0.5895
30	Logistic	0.6427	0.6413	0.6469
	k-NN	0.6303	0.7135	0.5177
	Tree	0.614	0.7556	0.4165
	R-forest	0.7025	0.7435	0.6477
	SVM	0.6803	0.809	0.504
	NN	0.6609	0.6478	0.588