Journal of Engineering Education Transformations

Abstract

Abstract— Educational Data Mining (EDM) is one of the trending areas in which various researchers are working for the betterment of the student’s performance. Predicting the students’ performance is considered as an important task in education sector and is of paramount importance as predicting the performance accurately may lead to great future of students by analyzing data properly. This article presents the review of 32 research articles which are from ACM, IEEE, Springer and Elsevier research database. This article analyzes these research articles based on number of research articles considered from research database, publication year, performance parameters, number of performance parameteres used by research articles, Data Mining Techniques, number of algorithms used by research articles, and dataset size. It is found that classification technique is used in EDM for analyzing students’ data and in classification technique, mostly employed algorithms are Random Forest, Logistic Regression, Decision Tree, Naïve Bays, Support Vector Machine and Knearest Neighbour. Generally the performance parameters such as accuracy, precision, recall and F-measures are used to decide the performance of the classification algorithms. This review article will be helpful to those researchers who are working in the EDM for predicting students’ performance for the dataset obtained from education sector.

Keywords—Data Mining, Educational Data Mining, Classification, Clustering, Association Rule