Predicting Student Engagement in Online Courses Using AI

Engagement among students is the most significant factor to be kept in mind while analyzing online learning statuses. Obviously, there must be some measurement on engagement, but for example self-reports, surveys, observations, and manual counts have failed to give up-to-the-minute accuracy. This work proposes a hybrid AI-based model that predicts levels of engagement using deep learning algorithms, in particular convolutional neural networks (CNNs) and Long Short-Term Memory (LSTM) networks. The dataset used consists of behavioral, physiological, and activity data from online learning platforms. The proposed model allows achieving 92.3 percent accuracy and illustrates the capability of AI in predicting student engagement. Moreover, a web-based tool was created to record real-time engagement so that interventions could be done at the time when students are showing signs of lack of engagement. This study outlines the importance of AI-derived engagement monitoring and how it can improve student learning results.