EduGage: A Multimodal Dataset and Benchmark for Sensor-Based Momentary Assessment of Engagement in Self-Guided Video Learning
Researchers developed a system to estimate learner engagement in video learning by combining multiple sensor signals like heart rate and eye movements. The method achieved high accuracy, significantly outperforming existing models. A new dataset, EduGage, is released to allow others to test and build upon this multimodal approach for assessing engagement in self-guided education.