Automatic Classification of Artifactual ICA-Components for Artifact Removal in EEG Signals

Table 2 Feature weight vector and test errors.

Feature	Feature weight	Test Error RT	Test Error ERP
Current Density Norm	0.342	0.141	0.488
Range Within Pattern	0.574	0.151	0.186
Mean Local Skewness 15 s	0.317	0.309	0.442
λ	0.569	0.177	0.144
8-13 Hz	-0.219	0.166	0.138
FitError	-0.286	0.424	0.640
Combined		0.089	0.147

Feature weights w_i for each feature x_i of the classifier H : ℝ⁶ ∋ x α sign(w^T · x + b) ∈ {-1, 1} with 1 ≙ Artifact and -1 ≙ Neuronal activity. Test error (MSE) for the 6 single features and for the combined classification for the RT experiment and the ERP experiment.

ISSN: 1744-9081