Predicting Student Attention And Engagement Through Eye-Tracking Using Machine Learning

This paper presents a comprehensive approach to predicting student engagement and learning behavior by analyzing eye-tracking data combined with behavioral analysis using camera-based recognition algorithms. The proposed system leverages WebGazer for eye-tracking to monitor gaze patterns and uses computer vision algorithms to recognize behaviors and emotional states. Feelings like engagement, frustration, or perplexity are identified using facial expression analysis. The system aims to provide real-time, actionable insights to students and educators by relating to emotional and concentration states by combining different modalities, thereby improving personalized learning strategies and educational outcomes.