Prediction of Student Graduation Time Using the Best Algorithm

Abstract

Data mining has a very important role in the world of education can help education institutions in predicting and making decisions related to student academic status. We use the NN, SVM and DT algorithms to predict the graduation time of academic students at one of the private universities in Indonesia. The results of this study indicate that the three models produce the accuracy of more than 80%, and the SVM model has an accuracy of 85.18% higher than the other two models. The results arising from this study provide important reference material for planning the future success of students and faculty in early warning to students in the future.