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

Journal of Big Data

Table 2 Results of performance metrics for arousal classification

Time	Classifier	Accuracy (%)	Precision (%)	Recall (%)	F1-score (%)
0 to 15 s	SVM	56.86	56.86	98.91	72.87
	LR	59.33	59.52	95.95	73.42
	DT	55.48	57.49	83.24	68.07
	KNN	52.95	59.95	63.15	61.05
	LDA	55.62	55.77	97.12	70.45
15 to 30 s	SVM	63.39	63.88	99.56	77.87
	LR	64.54	64.53	99.26	78.19
	DT	63.35	65.10	92.78	76.38
	KNN	59.81	63.91	85.60	73.31
	LDA	63.12	63.32	99.50	77.39
30 to 45 s	SVM	73.75	73.77	99.50	84.73
	LR	63.82	64.39	97.41	77.53
	DT	64.06	64.06	100	78.09
	KNN	58.88	63.73	83.02	72.11
	LDA	63.28	63.46	99.26	77.42
45 to 60 s	SVM	63.59	63.67	99.50	77.66
	LR	64.82	64.49	77.05	77.95
	DT	64.53	64.44	99.96	78.26
	KNN	63.35	69.35	99.75	77.66
	LDA	63.12	63.32	99.50	77.39