Analysis of Learner's Behavior in Online Video Course Based on Eye Movement Tracking

The interdisciplinary course combines two or more academic disciplines from different fields, requiring learners to know many areas. In e-learning, the students may have different backgrounds and knowledge levels. It is critical for a teacher to understand the e-learners' difficulties and interests of various course concepts and their preferred learning styles. As eye movement is essential to recognize the students' learning behavior during the learning process, this study proposed a knowledge unit personalized classification method based on eye movement tracking in order to assist teachers to measure knowledge units' difficulty and importance for the students of different knowledge levels. The course teaching content with knowledge units is prepared in the logical and pedagogical organization. Then, we calculate the time spent on each knowledge unit based on eye fixations. A classification approach based on student learning time analysis was adopted to categorize the difficulty and importance level of knowledge units for different students. Finally, we used learners' knowledge levels based on the questionnaire to evaluate this proposed approach. The evaluation results show the effectiveness of the approach.