Framework to Identify Academic Affect States of Students Observed During Online Learning

Online learning has become prevalent and inevitable in modern education. Understanding students’ emotions is essential for enhancing the efficacy of online learning environments. This thesis introduces an Academic Emotion Recognition Framework that uses webcam videos for unobtrusive data collection and analyzes facial expressions, head pose, and eye-gaze behavior to identify student emotions during online learning. The framework integrates multiple visual cues to define academic-specific emotional expressions. The outcomes of this research are anticipated to facilitate future studies in the academic emotion recognition domain, providing framework for developing advanced models and technologies to enhance the overall effectiveness of online learning platforms.