Table 2.

Emotion prediction performance of machine learning algorithms in the time domain.

Class	Index	LR	NB	SVM	RF	KNN	CNN-LSTM	SNN	SBP-SNN
Fearful and Neutral	Acc. (%)	28.62	47.67	71.94	76.24	66.74	87.50	76.72	81.82
	F1-Sc.	0.28	0.55	0.48	0.72	0.58	0.75	0.74	0.84
Sad and Neutral	Acc. (%)	57.14	52.44	66.71	42.96	56.22	75.00	78.19	84.09
	F1-Sc.	0.39	0.39	0.73	0.26	0.52	0.72	0.80	0.86
Happy and Neutral	Acc. (%)	33.33	47.36	72.41	66.75	60.11	81.25	75.48	81.82
	F1-Sc.	0.25	0.25	0.64	0.68	0.58	0.72	0.76	0.83
Four classes	Acc. (%)	–	55.63	62.62	53.37	56.05	61.00	53.88	63.86
	F1-Sc.	–	0.44	0.57	0.43	0.50	0.59	0.56	0.66