EEG-based Binary Classification of Brain State of Activities Level Using a Single-Sensor Headset

Electroencephalography (EEG) is a non-invasive procedure that is used to measure and record the electrical activity of the brain. EEG uses scalp electrodes to measure brain activity, commonly used to diagnose epilepsy, brain tumors, sleep disorders, and stroke. The goal of this paper is to classify the brain state activities of the individual based on the EEG readings obtained through the Neuro Sky Headset. In this paper, several classification methods were utilized in order to accurately predict the subject’s brain state activities, including logistic regression, linear SVC, nonlinear SVC, and gradient boosting. The analysis in terms of accuracy, precision, recall, and F1-score is done to figure out performance of these methods on brain activity data. The gradient boosting algorithm yielded a maximum accuracy of 81%.