Classification of EEG data for human mental state analysis using Random Forest Classifier

Abstract

Brain computer interface (BCI), has been one of the most popular domains in computing in the recent years. BCI is a pathway which allows communication between computers and the human brain. We acquire real time EEG data with the device, Neurosky Mindwave Mobile, which uses a single dry electrode. Experiment for acquisition of data is carried on 40 subjects (33 male and 7 female). Feature extraction of EEG signals are done by statistical measures such as mean, standard deviation, maximum and minimum amplitudes. In this paper we explore the approach of ensemble learning with classifiers such as random forest classifier to build a BCI model to predict mental states as concentration and meditation. Analysis and results of our proposed model shows an accuracy of 75% using the above methodologies. This model is further implemented in the field of Internet of Things (IoT), for the application of home automation.