Predicting Academic Performance of Students from the Assessment Submission in Virtual Learning Environment

The growth of online learning platforms has transformed the educational world, and the research community is now more interested than ever in discovering optimal ways to exploit online education repositories, including students’ interactions with such platforms. Early prediction of student academic success has been a prevalent study issue, and multiple studies have articulated several measures and predictors for such analysis. In this work, we hand-engineer numerous factors relevant to a student’s first evaluation in a course using data obtained from the Open University in the United Kingdom. A series of typical machine learning methods runs an array of trials on two feature sets, incorporating demographics with interactions and assessment data and ignoring demographics. The outcomes are examined to find key characteristics and determinants of student success. Such research is expected to aid in developing appropriate educational policies for an active layer of student support and intervention. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.