Student Attitude Identification towards E-learning Course Based on Biosensor Information

Providing attractive and interesting e-learning course materials as well as delivering normal lectures in class are important issues in an e-learning society. Knowing about this issue, first, it is essential to investigate student interest about topics and types of elearning courses delivered by measuring the students‘ responses. Thus, this study engages in evaluation of students‘ responses initiated by two distinct e-learning materials; one is characterized by interactive material and the other by non-interactive material based on biosignals—i.e., Blood Pressure (BP), Galvanic Skin Response (GSR), and Electrocardiogram (ECG) signals. Use of a multilinear principal component analysis (MPCA) was proposed in this study to extract the representative features of student responses. Finally, the classification was performed by a support vector machine (SVM) to discriminate between responses of students.