Gender Recognition using EEG during Mobile Game Play.

In recent years, due to an exponential increase in mobile game users, automatic recognition of personal traits i.e., gender and user identity is gaining significance. Personal trait recognition also plays a vital role in information security. Traditional approaches for personal trait recognition, including vision and voice-based schemes, are subjected to privacy constraints. To address these challenges, physiological signals such as electroencephalogram (EEG) and electrocardiogram have been utilized in information security for bio-metric and personal attribute identification. Herein, we propose a new technique for EEG-based gender recognition during mobile game play. Towards this, EEG data of 38 users are recorded using the Muse headband while playing a mobile game named Traffic Racer. Different time and frequency-domain features are extracted from the acquired EEG data. Three classification algorithms, i.e., K-nearest neighbor (KNN), random forest, and the Naive bayes, are used to train and validate the proposed model. In particular, KNN achieved an accuracy of up to 97.36% in gender recognition, which shows the efficacy of the proposed approach when compared with other techniques.