Privacy-preserving on-screen activity tracking and classification system for E-learning

In the evolving field of e-learning, addressing student engagement and academic behavior tracking challenges is crucial. The project "Privacy-Preserving On-Screen Activity Tracking and Classification System for E-Learning" offers a secure, automated framework for monitoring without compromising privacy. Utilizing anonymized screen captures analyzed through deep learning, particularly Convolutional Neural Networks (CNNs), it classifies activities such as Reading, Watching Videos, and Browsing. This approach avoids storing personal data and ensures ethical compliance. Experimental results indicate an accuracy of 90-95% in activity classification, outperforming traditional monitoring systems. The system aims to enhance E-learning Analytics and Privacy-Aware AI by offering insights into student engagement while protecting user data.