EEG-Based Emotion Classification for Verifying the Korean Emotional Movie Clips with Support Vector Machine (SVM)

Table 8

Emotion classification accuracy from different channels based on support vector machine using grid search.

Emotion^1)	Frequency bands	Hyperparameter	All channels^2)	Selected channels^3)
All emotions; (A, E, F, H, N, S)	All bands		71.86
	Alpha + Beta		79.97
			84.11
5 emotions; (A, E, F, H, S)	All bands		80.85
	Alpha + Beta		89.28
			91.20

Note: : 1/Number of features x variance of data). ¹⁾Emotion: A (anger), E (excitement), F (fear), H (happiness), N (neutral), S (sadness). ²⁾All channels: AF3, AF4, F7, F8, F3, F4, FC5, FC6, T7, T8, P7, P8, O1, and O2. ³⁾Selected channels: AF3, AF4, F3-F4, F7, F8, P7, P8.