Artificial Intelligence based eLearning Model to Increase Student Engagement

Many web applications provide online courses for students to upskill their knowledge. This type of learning system is unique for the students but associates several challenges like lack of motivation to complete the course, higher dropout rate, and low engagement. The research paper aims to create a model that focuses on finding low-engagement students and the learning activities that play a significant role in increasing user engagement. Machine learning (ML) algorithms like Decision Tree, Gradient-Boosted Tree, and Naive Bayes are applied to the dataset provided by Open University (OU) to find low-engagement students and the impression of certain activities on the performance of students. The dataset’s input variables are the highest level of education, the assessment score, the final outcome, and the number of clicks on Virtual Learning Environment (VLE) activities such as oucollaborate, glossary, oucontent, subpages, forumng, homepage, resources, dataplus, and URL. The degree of student engagement, whether high or less, is the output variable. The result showed that the gradient-boosted classifier performs better in terms of accuracy as compared to other tested models to determine the low-engagement students.