An Evolutionary Online Framework for MOOC Performance Using EEG Data

Abstract

Massive Online Open Course (MOOC) is a scalable, free or affordable online course which emerged as one of the fastest growing distance education platforms in the past decade. One of the biggest challenges that threatens distance education is abnormality in the overall level of consciousness of students while they are taking the course. In this paper, an evolutionary online framework was proposed to improve the performance of MOOCs via noninvasive electro-physiological monitoring methods such as electroencephalography (EEG). Based on the proposed platform, EEG signals can be recorded from users while they are wearing any EEG headsets. EEG measures a brain's spontaneous voltage fluctuations resulting from ionic current within the neurons of the brain via multiple electrodes placed on the scalp. A total of eleven extracted features from EEG signals were employed as the inputs of the evolutionary classification algorithm to predict two classes of confused and not-confused for each individual. An accuracy of 89 % was considered significant enough to suggest that there is difference in the EEG signals of individuals with confusion versus not-confused individuals.