REAL TIME ATTENTION SPAN TRACKING IN ONLINE EDUCATION

Abstract: Over the last 10 years, e-learning has changed the way students study by offering them access to high-quality instruction whenever and wherever they need it. However, understudies regularly become preoccupied for a variety of reasons, all of which have a negative impact on their learning ability. Many experts have attempted to improve the nature of online education, but we need a comprehensive solution to this problem. This research aims to present a method that uses a camera feed and a mouthpiece contribution to monitor student's continuing attention levels during online classes. Throughout this review, we look at several photo handling strategies and AI algorithms. We suggest a framework that utilises five specific non-verbal features to calculate the understudy's consideration score throughout computer-based tasks and generate continual input for both the understudies and the association. We can use the generated data as a heuristic value to investigate the general exhibition of understudy as well as the speaker's exhibiting guidelines.

Keywords: Artificial Intelligence, Attention, Blink rate, Drowsiness, Eye look following, Emotion arrangement, Face acknowledgment, Body Posture assessment, Noise recognition.