Improving accuracy of students' final grade prediction model using PSO

The basic idea of Educational data mining is to extract hidden knowledge in educational field using data mining techniques. The suggested system for the purpose of predicting student performance applied in this study is carried out in two major phases. In the first phase, the feature space is searched to reduce the feature numbers and prepare the conditions for the next phase. This task is carried out using several dimension reduction techniques. Afterwards, a subset of features is chosen for the classification phase. Experimental results demonstrated that the proposed technique based on PSO could improve the accuracy performance and achieve promising results with a limited number of features. The present study will also promote the future investigation of evaluating the proposed approach over other student performance datasets and test other optimizations and classification algorithms.