Classification of emotional states from electrocardiogram signals: a non-linear approach based on hurst

Table 2 FVS based analysis for emotion classification

Feature	Classifier	Neutral	Happy	Sad	Fear	Surpris	Disgus	Average
		(%)	(%)	(%)	(%)	e (%)	t (%)	(%)
Hurst	Regression tree	54.72	61.90	67.12	65.07	60.96	76.03	64.30
	Bayesian classifier	41.22	48.30	46.58	78.10	36.99	92.47	57.27
	KNN (K = 6)	58.78	62.58	64.38	65.06	64.38	82.19	66.23
	FKNN (K = 11)	55.40	59.18	69.17	63.01	66.43	78.76	65.33
Hurst_skewness	Regression tree	61.49	64.63	78.08	77.39	73.97	82.19	72.95
	Bayesian classifier	29.05	79.59	53.42	71.91	73.97	43.84	58.63
	KNN (K = 8)	59.45	62.58	69.86	85.61	74.65	79.45	71.93
	FKNN (K = 8)	66.21	61.90	76.71	80.13	77.39	85.61	74.66
Hurst_kurtosis	Regression tree	68.91	64.62	84.24	91.09	84.25	90.41	80.59
	Bayesian classifier	44.59	70.74	61.64	79.45	36.30	100	65.45
	KNN (K = 10)	67.56	67.34	80.13	91.09	83.06	95.20	80.81
	FKNN (K = 10)	72.97	65.30	85.61	90.41	86.30	89.72	81.72

ISSN: 1475-925X