Pilot Anxiety Detection through Brain Signal Using Naïve Bayes Method

Air transport is vital for Indonesia, which consists of more than 17,000 islands with vast airspace. In addition to supporting connectivity and mobility between regions, air transportation also plays a role in economic growth and tourism. Despite these benefits, air transport faces risks and challenges, especially related to human factors such as fatigue and stress on pilots. Therefore, real-time monitoring of the pilot's physiological and psychological condition during flight is necessary to prevent errors and accidents. This study aimed to detect pilot anxiety from electroencephalogram (EEG) signals using the Naïve Bayes method. EEG signal data was obtained from 16 prospective pilot subjects who participated in flight simulations under normal conditions. The data is processed using band pass filters, feature extraction, and feature selection. The features used are the amplitude and frequency of EEG waves in each electrode channel. The selected features were then classified using the Naïve Bayes approach resulting in an accuracy of 96.1%. This research contributes to the field of neuroergonomics, and can be useful for improving pilot safety and well-being.