Optimal mental task discrimination for brain-computer interface

A Support Vector Machine (SVM) classification method for data acquired by EEG recording for brain/computer interface systems is here proposed. The aim of this work is to evaluate the SVM performance in the recognition of a human mental task, among others. A prerequisite has been the developing of a system able to recognize and classify the following four tasks: thinking to move the right hand, thinking to move the left hand, performing a simple mathematical operation, and thinking to a nursery rhyme. The data set exploited in the training and testing phases has been acquired by means of 61 EEG electrodes and consists of 4000 time series. These time data sets were then transformed into the frequency domain, in order to obtain the power frequency spectrum. In such a way, for every electrode, 128 frequency channels were obtained. Finally, the SVM algorithm was used and evaluated to get the proposed classification. Different choices of electrodes have been considered: we found that analysing only a subset of electrodes we can get better results than considering all the 63 electrodes.