ECG-based affective computing for difficulty level prediction in Intelligent Tutoring Systems

Abstract

Intelligent tutoring Systems (ITS) have emerged as an attractive solution for providing personalised learning experiences on a large scale. Traditional ITS are able to adapt the learning process according to the capabilities and needs of their users, but lack the capability to adapt to their affective/emotional state. In this work, we examine the use of electrocardiography (ECG) signals for detecting the affective state of ITS users. Features, extracted from ECG signals acquired while users undertook a computerised English language test, were used for the prediction of the self-reported difficulty level of the test's questions. Supervised classification experiments demonstrated the potential of this approach, achieving a classification F1-score of 61.22% for the prediction of the self-assessed difficulty level of the questions.