A Hybrid Intelligent System Framework for the Prediction of Heart Disease Using Machine Learning Algorithms

Table 8

10-fold CV classification performance of different classifiers on selected features by mRMR FS algorithm when .

Predictive model	Classifiers performance evaluation metrics
	Turning parameters	Accuracy (%)	Specificity (%)	Sensitivity (%)	MCC	AUC (%)	Processing time (s)
Logistic regression	C = 1	74	82	66	74	74	2.313
	C = 10	75	82	67	74	75	2.352
	C = 100	78	88	67	78	79	2.159
K-nearest neighbor	K = 1	57	57	58	57	63	1.784
	K = 3	56	56	55	56	55	1.742
	K = 7	62	62	61	62	65	10.144
Artificial neural network	16	63	67	58	62	66	30.802
	20	47	4	98	51	50	23.483
SVM (kernel = RBF)	C = 100, = 0.0001	77	88	65	76	77	60.589
	C = 10, = 0.001	66	71	60	65	67	59.132
SVM (kernel = linear)	C = 10, = 0.0001	58	23	70	60	59	12.567
	C = 100, = 0.0001	70	100	35	68	71	10.179
Naive Bayes	—	84	90	77	83	84	1.596
Decision tree	100	57	55	60	58	57	1.902
	50	60	54	67	60	61	1.831
Random forest	100	66	69	62	66	65	1.121
	50	67	70	62	66	68	2.220