Comparison of Naive Bayes Classifier and C4.5 in Predicting Student Study Period

Abstract

A department of an Islamic State University in Indonesia has an average graduation presentation on time of 13.5%. This is very worrying, it has an impact on various things. Therefore, the department will find it difficult to obtain optimal assessment values. The purpose of this study was to analyze the data of active students to gain knowledge of what caused and what factors influenced the student's study period. This study proposes a classification model that can be used to predict student study periods using the Naive Bayes Classifier and C4.5 algorithms. In the determinant analysis phase, this work discovers that there are several attributes of active students who influence on the student's study period such as gender, GPA, college entry scheme, tahfidz, etc. In the testing phase, this work conducted that Naive Bayes Classifier has a better accuracy rate than C4.5.