PEMODELAN PREDIKSI TINGKAT KELULUSAN MAHASISWA DENGAN PENDEKATAN ALGORITMA NAÏVE BAYES

Abstract

One important factor in assessing the accreditation of an institution or higher education institution is graduation. It is important to know the graduation predictions of Informatics Engineering Study Program students at the University of August 17, 1945 in order to identify students who did not graduate on time from the start. Prediction of student graduation can be done using data mining, especially the classification method. In this study, the Naive Bayes Algorithm is used as a classification method. The training data consisted of 120 alumni of the Informatics Engineering Study Program class of 2017, while the test data consisted of 30 students of the 2018 class. The attribute data used for this research were Social Sciences Semester 1-4, Credits, GPA, and student graduation status. This research is expected to provide information about predictions of student graduation on time and provide input for tertiary institutions for future improvement. The results of this study include the results of applying the Naïve Bayes algorithm to this graduation prediction system that can be used to estimate whether students will graduate on time or not. Students can easily self-introspect by accessing this website page online without having to go to the building. academic bureau, the results of this study have an accuracy rate in predicting student graduation that is 77%, 77% precision level, and 78% recall time.