EEG Signal Classification Using Manifold Learning and Matrix-Variate Gaussian Model

Table 3

Cross-validation performance results for different algorithms in Exp.2.

Feature Extraction	Subj.1 (% m, )	Subj.2 (% m, )	Subj.3 (% m, )	Average (%)
None	81.67 m = 1	56.67 m = 1	58.33 m = 2	65.56
LDA	78.89 m = 1,	58.33 m = 2,	53.33 m = 2,	63.52
2DLDA	79.89 m = 2,	58.50 m = 2,	55.17 m = 2,	64.52
DLPP	85.00 m = 1,	57.50 m = 1,	52.50 m = 1,	65.00
2DDLPP	85.44 m = 3,	54.33 m = 1,	57.67 m = 1,	65.81
B2DDLPP	87.22 m = 1,	60.00 m = 2,	64.17 m = 1,	70.46

For each method and each subject, optimal related to FBCSP’s output and the optimal dimension () are presented.