<![CDATA[This study aims to predict student graduation at Universitas Nahdlatul Ulama (UNU) Lampung using academic data with the Decision Tree C4.5 algorithm in RapidMiner. This research is based on the problem that some students graduate late and academic data is not fully used for decision-making. The method used is a quantitative approach with an experimental design following the CRISP-DM stages, which include business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The data used consists of 389 student records from four study programs. The variables include GPA, semester GPA, gender, and study program, while graduation status is used as the target variable. The results show that GPA is the most important factor affecting graduation. Students with GPA ≤ 3.00 tend to graduate late. The model produced an accuracy of 85.34%, precision of 87.50%, and recall of 97.03%. Therefore, it can function as an early warning mechanism to support academic programs in increasing on-time graduation rates.]]>
Copyrights © 2026
