Predicting student performance: a classification model using machine learning algorithms

With the increasing availability of educational databases, extraction of interesting patterns and relationships from such data becomes extremely attractive and challenging. Discovering implicit patterns related to student performance is potentially helpful to enhance student achievement. In this paper, a student performance prediction model is generated utilising machine learning algorithms. The central idea is that identifying the dominant features that affect student performance results in generating an effective student performance prediction model. In order to achieve this goal different feature selection approaches are considered. The reported experimental results indicated that the effectiveness of student performance prediction model is significantly affected by the dimensions featured in the considered dataset.