Fusion of Motif- and Spectrum-Related Features for Improved EEG-Based Emotion Recognition

Table 2

Comparison of different feature selection algorithms and number of features (nof) for benchmark feature set.

Feature groups	Valence		Arousal
	BACC	nof	BACC	nof
ANOVA	0.5490	9	0.5316	8
mRMR	0.5404	3	0.5281	4
RFE	0.5531	3	0.5318	15