A classification method for prediction of qualitative properties of multivariate EEG-P300 signals

A classification method is used to predict the qualitative properties of a subject's mental state by extracting useful information from the highly multivariate non-invasive recordings of brain activity. In this paper, an application of a classification method entailing time-series EEG signals with backpropagation neural networks is presented. To test the improvement in the EEG classification performance (i.e., classification accuracy and transfer rate) with the proposed method, comparative experiments were conducted with other classifier which is Bayesian Linear Discriminant Analysis. Finally, the promising results reported that up to 97% average classification accuracy and 42.4% improvement of maximum average transfer rate is achieved.