Data Mining Techniques in EDM for Predicting the Pupil’s Outcome

Abstract: In recent era, growth of higher education has increased massively. Many new institutions and graduation schools are being established by both the private and government sectors for the growth of education and welfare of the students. Each institution aims at producing higher and exemplary graduation rates by employing various teaching and grooming methods. But still there are certain cases of unemployment that exists among the medium and low risk students. This paper aims to describe the use of data mining techniques to improve the efficiency of academic performance in the educational institutions. Various data mining techniques such as clustering, decision tree, association and rule induction, nearest neighbors, neural networks, genetic algorithms, exploratory factor analysis and stepwise regression can be applied to the higher education process, which in turn helps to improve pupils‟ performance. These approaches fit to provide a model to the problem domain that takes place in the educational systems.

Keywords: Data Mining, KDD, EDM, ANN, Decision Trees, Association, Clustering.