Classifying subjective emotional stress response evoked by multitasking using EEG

摘要

In this fast pace world, individuals are expected to perform and complete multiple jobs simultaneously which induces various levels of stress among persons. Though positive stress can contribute effectively towards success, its overload could cause serious health problems. Therefore, it is of paramount importance to understand and differentiate good stress from bad stress for each person. Even though a number of studies reported in literature have identified stress/emotion using brain signals, investigation of stress with user's emotional response is rarely explored. In this study, our aim is to understand human stress based on emotional response of the users, using band power features extracted from Electroencephalogram (EEG) signal. We are using multitasking framework to evoke the subjective emotional stress response. Using Support Vector Machine technique that effectively differentiates good and bad stress from the relaxed state of mind, an average recognition accuracy of 77.53% is obtained in the three level stress classification.