Decision Tree-Based Student Graduation Prediction System at the Faculty of Technology, University of Battuta

Abstrak

On-time graduation is an important indicator of academic quality in higher education. However, the Faculty of Technology at Battuta University still faces the challenge of low on-time graduation rates among students. This study designed a graduation prediction system based on the C4.5 Decision Tree Algorithm by utilizing the academic data of students from the 2022–2024 batch, including GPA, IPS, failed courses, academic leave, status, and attendance. The method used is a quantitative approach with data mining classification techniques, and the system is implemented web-based using PHP and MySQL. The results show that the C4.5 algorithm is capable of accurately predicting student graduation potential and producing classification rules that are easy to understand. This system can help academic advisors and study programs detect students at risk of graduating late at an early stage so that appropriate follow-up actions can be taken.