Assessment of distinction sensitive learning vector quantization weighted common spatial pattern features for EEG classification in Brain Computer Interface

Abstract

Common Spatial Pattern (CSP) is a method commonly used to find spatial filters for data classification in multichannel EEG-based Brain Computer Interface (BCI) systems. In the present study, a novel CSP sub-band feature selection has been proposed based on the discriminative information of the features. Besides, a DSLVQ-based weighting of the selected features has been considered. Then, the selected and weighted features have been classified using an SVM classifier. Finally, the performance of the suggested method has been compared with the basic CSP on EEG data on 5 subjects, from BCI competitions datasets. The results show that the proposed method outperforms the basic CSP algorithm by %7.3 on the average.