A comparative analysis of machine learning methods for emotion recognition using EEG and peripheral physiological signals

Journal of Big Data

Table 1 Results of performance metrics for valence classification

Time	Classifier	Accuracy (%)	Precision (%)	Recall (%)	F1-score (%)
0 to 15 s	SVM	54.53	55.52	92.91	69.27
	LR	56.50	57.42	83.81	68.17
	DT	52.98	55.93	69.40	61.97
	KNN	53.90	57.93	62.12	59.95
	LDA	55.00	64.48	65.77	65.30
15 to 30 s	SVM	63.34	63.34	100	77.67
	LR	62.17	63.42	96.67	76.60
	DT	64.06	64.063	100%	78.09
	KNN	55.55	63.67	71.21	67.24
	LDA	57.34	67.72	69.79	68.72
30 to 45 s	SVM	63.43	63.43	100	77.62
	LR	63.12	64.05	96.67	77.05
	DT	61.60	64.86	100	78.09
	KNN	70.41	70.20	93.23	83.91
	LDA	56.25	67.87	66.05	67.05
45 to 60 s	SVM	63.59	63.35	99.50	77.73
	LR	63.12	64.05	96.67	77.05
	DT	64.60	64.66	100	78.09
	KNN	57.68	64.04	77.49	70.11
	LDA	56.25	68.85	64.18	66.34