IoT-Enhanced Learning Environment Optimization and Student Outcome

Abstract

This proposed system leverages Internet of Things (IoT) technology to enhance the learning environment in educational settings through two synergistic techniques. Firstly, a search-based optimization algorithm, driven by a genetic-based approach, is implemented for scheduling courses and faculty within each department to improve overall student performance and departmental percentages. Secondly, a classification task is performed to predict student outcomes, employing Neural Networks (NN) including ResNet 50, ResNet34, and a hybrid ResNet34 and ResNet50 model. The classification is based on eye-gaze monitoring during active student engagement in class, using input video samples as training and testing datasets. The system integrates optimization, activity monitoring, and classification to create a comprehensive approach aimed at improving the overall learning environment and student outcomes in educational institutions.