A CNN based model for Identification of the Level of Participation in Virtual Classrooms using Eye Movement Features

As the widespread Internet connection is expanding, the transition from traditional classroom learning to virtual learning is now easier than ever. Online education has developed all over and replacing traditional learning. The ability to precisely monitor user behavior and understand where the pain spots are in the learning process is one of the benefits of the change to online learning. Therefore, this research will provide a benefit to Online learning education. Here, Fixation recognizable proof, which involves confining and identifying Fixation and saccades in eye-tracking conventions, is a vital piece of eye movement information handling that can altogether affect more elevated level examinations. Fixation distinguishing proof techniques, then again, are normally talked about casually and seldom analyzed. The work specified gives a scientific classification of fixation distinguishing proof calculations that groups calculations as per how they utilize spatial and fleeting data in eye-following conventions. Utilizing this scientific categorization, the Adaptive algorithm that is suggestive of one-of-a-kind classes in the scientific classification and is based on regularly utilized strategies. Then, at that point, utilizing a bunch of subjective rules, we investigate and look at this algorithm. Here, CNN has been utilized for face information registering and Adaptive calculation for eye fixation. The after effects of these correlations have interesting ramifications for how algorithms will be utilized later on. Providing an expected results on adapting for eye contact lenses and kajal strengthens the research work. An alert and email system that notifies the participant if there is a lack in focus during the online class is proposed.