COMPARATIVE STUDY OF SUPERVISED LEARNING FOR STUDENT PERFORMANCE EVALUATION

Data mining technique can help bridging this knowledge gap in the higher educational system. The data mining methodology helps for betterment of efficiency in educational institutions. The data mining approach such as classification, association rule mining, clustering, prediction, etc.is used to improve students' achievement. It helps in their life cycle management and assist in the selection of the course. The classification is an important data mining task and it can be applied very effectively in educational data. In this study, the application of a classification technique in education data mining is focused. The comparative study was conducted for the prediction of a student's academic performance based on social variable, pervious exam grades and other attributes related tothe performance of students. The J48, Naïve Bayes, Bayes Net, Back propagation network and Radial Basis Function Network classification techniques were considered for the experiment. The result revealed that correctly classify instance percent is 100% of Radial Basis Function Network and its high compare to other classifier.