Comparison of Data Mining for Classifying Student Graduation Levels Using Naive Bayes, Decision Tree, and Random Forest Methods (Case Study of The Undergraduate Program at Mitra Indonesia University)

Abstract

This study aims to apply data mining techniques to classify student graduation rates in the Undergraduate Program at Mitra Indonesia University. The methods used in this study include Naive Bayes, Decision Tree, and Random Forest. The data used includes student academic data, such as grades, attendance, and other demographic information. The research steps include data collection, data cleaning, data analysis, and the application of data mining algorithms. The results of the study show that the Random Forest method provides the highest accuracy compared to Naive Bayes and Decision Tree in predicting student graduation rates. The Random Forest method achieved an accuracy of 85%, while the Decision Tree achieved 80%, and Naive Bayes achieved 75%. These findings are expected to help Mitra Indonesia University identify students at risk of not graduating on time, so appropriate interventions can be provided to improve graduation rates