A Framework on Performance Analysis of Mathematical Model-Based Classifiers in Detection of Epileptic Seizure from EEG Signals with Efficient Feature Selection

Table 7

Values of TP, TN, FP, FN, and MSE for Haar, dB4, and Sym 8 wavelet features in various classifiers without feature selection.

Wavelets	Classifiers	TP	TN	FP	FN	MSE
Haar	Linear regression	57	66	34	43	6.48E − 05
	Nonlinear regression	58	57	43	42	8.38E − 05
	GMM	68	75	25	32	3E − 05
	K-NN	55	87	13	45	7.44E − 05
	SVM (linear)	62	72	28	38	4.16E − 05
	SVM (polynomial)	65	62	38	35	5.35E − 05
	SVM (RBF)	69	85	15	31	2.14E − 05
dB4	Linear regression	70	57	43	30	0.000143
	Nonlinear regression	55	66	34	45	8.29E − 05
	GMM	63	84	16	37	2.98E − 05
	K-NN	61	57	43	39	7.46E − 05
	SVM (linear)	72	57	43	28	5.53E − 05
	SVM (polynomial)	63	53	47	37	0.000199
	SVM (RBF)	53	55	45	47	0.000194
Sym8	Linear regression	54	64	36	46	0.000138
	Nonlinear regression	59	87	13	41	3.56E − 05
	GMM	55	73	27	45	0.000102
	K-NN	55	81	19	45	6.7E − 05
	SVM (linear)	71	57	43	29	5.78E − 05
	SVM (polynomial)	57	83	17	43	4.44E − 05
	SVM (RBF)	90	95	5	10	8.89E − 06