Predicting Student Academic Performance Using Temporal Association Mining

Abstract

Predicting the performance of a student is of great concern to the higher education managements. In the real world, predicting the performance of the students is a challenging task. Many of the well known technical colleges are successful as they have meritorious students and faculty with them and a foot proof system working for them to grow continuously. The primary goal of Data mining in practice tends to be Prediction and Description. In this paper, we proposed a method to predict the intra-year Academic Performance of the student using the historic data. The idea is to identify existing patterns in the historic data, and maintain a database for it. Comparing the current performance of a student with the existing patterns, we predict the possible performance of the student in the future. In the process, we identify any new patterns that we come across. For the purpose of identifying the interestingness of the pattern we look into the percentage of increase or decrease we make, appropriate methodologies are used.