Improving the Performance of P300-Speller with Familiar Face Paradigm Using Support Vector Machine Ensemble

Abstract

P300-speller is a communication style based on Brain-computer interface (BCI) which allows users to input characters by electroencephalography (EEG) signals. In the past few years, there are various studies on P300-speller paradigm and classification algorithm. However, the accuracy and bit rates are not yet satisfied for our daily life. In order to improve the performance of the P300-speller, we designed an experiment in which support vector machine ensemble for P300-speller with familiar face paradigm was used. Seventeen subjects participated in the experiment and achieved a good classification accuracy. The results showed that support vector machine ensemble enhanced the performance of P300-speller with familiar face paradigm.