Computer Interfaced Effective Analysis Using Morpheme on Brain Signals

Abstract

Brain Computer Interface (BCI) has become hugely big in the present backdrop especially in the research arena because of its efficiency and effectiveness in the results it produces. A wireless headset, which helps in sensing the brain signals with the help of sensors available in it, is taken for the study. This headset has P300 event-related potential which helps in obtaining the Electroencephalogram (EEG) signals with ease. BCI finds its applications in various walks of life, particularly in medical field, field of knowledge and practical implementations. In this paper, a versatile method is proposed for processing EEG signals. EEG interpretation helps in the diagnosis of disorders in the brain and also in detecting if the information given by the subject is true or false. It uses classifiers such as Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes classifier and K-Nearest Neighbor (KNN) algorithm for the purpose of classifying the data that the headset produces. All four algorithms are used and compared to find out the best suitable for the purpose of classification when brain signals are taken into consideration.

Keywords: Electroencephalogram (EEG), P300, Classifier, Brain Computer Interface (BCI)