A systematic literature review on predicting students academic performance by using data mining techniques

There has been a lot of interest in education on the prediction of students academic performance. The enormous increase in educational data offers the chance to gather data that can be used to assess the effectiveness of teachers, anticipate student dropout rates, predict overall academic achievement, revise the material to better suit the requirements of students, and much more. However, the lack of a mechanism in place to predict students' academic performance is still a concern in Malaysia. The research on existing prediction techniques is still inadequate and very few studies that have been done on the Malaysian context, especially that contribute to students' academic performance. Given the scarcity of research on existing prediction techniques in Malaysia context, a detailed literature review on employing data mining techniques to predict student performance is suggested. The primary goal of this article is to provide a thorough overview of data mining approaches to predict students academic performance, as well as how various prediction techniques aid in determining the most significant students attributes which contribute to students performance. The findings of this paper offer an insight of the implementation of data mining in a specific dataset, and it summarizes the prediction algorithm with highest accuracy and attributes with significant contributions to students academic performance.

Actions (login required)

View Item