Students performance prediction using KNN and Naïve Bayesian

Abstract

Data mining techniques is rapidly increasing in the research of educational domains. Educational data mining aims to discover hidden knowledge and patterns about student performance. This paper proposes a student performance prediction model by applying two classification algorithms: KNN and Naïve Bayes on educational data set of secondary schools, collected from the ministry of education in Gaza Strip for 2015 year. The main objective of such classification may help the ministry of education to improve the performance due to early prediction of student performance. Teachers also can take the proper evaluation to improve student learning. The experimental results show that Naïve Bayes is better than KNN by receiving the highest accuracy value of 93.6%.