A Data Driven Model for Predicting the Academic Performance of Students Employing ANN-PSO Hybrid Approach

Abstract

Predicting student performance is critically important to foster a positive and conductive learning environment and to maximize educational potential. However, student performance is a very challenging metric to forecast owing to the wide and diverse set of factors which may affect student performance. By developing an accurate data driven model, we can identify potential drop outs and take countermeasures to evade such scenarios. Additionally, different additional inputs can be tailor made to cater to the specific needs of the different categories of students. The aim however, is not to discriminate among students but rather come up with strategies to compensate for lacuna in the prevailing teaching methods. This paper presents a data driven approach to predict student academic performance based on the deep neural network model. Data optimization has been done employing the principal component analysis. The performance of the proposed system has been evaluated in terms of the prediction accuracy.