Analyzing Student Engagement to Enhance the Online Teaching–Learning Environment

Research in educational data science has recently been abundant and pertinent. In the past two years of the pandemic, the concept of online learning has significantly increased flexibility in access to education. Although teaching in the virtual world is a significant educational and technological issue. Designing successful online learning environments that can suit the need of students’ expectations is a field that is always evolving. However, such a task yields substantial rewards. Mainly based on learning analytics (LA) and educational data mining (EDM) communities, virtual platforms are emerging to generate information regarding learning models. Analyzing student performance in an online learning environment at an early stage of course commencement may yield to categorize students into advanced and slow learners which can then be used to provide a customized learning environment for better results. In this paper, we have proposed a method to analyze student engagement. It considers online activities on a learning management system (LMS). Student engagement in an online environment like user activities, and behavioral and temporal data are taken into consideration. Based on these features, our proposed method has compared the result of student engagement using the fuzzy miner algorithm and the sequential pattern mining algorithm and categorized into active, passive, and disengaged which helped us to identify advanced or slow learners. Predicting student performance will also help in resolving the student retention issue which is quite a big problem in an online environment.