<![CDATA[This study aims to apply data mining techniques to predict student graduation using the C4.5 Decision Tree algorithm in the Information Systems Study Program, Faculty of Science and Technology, Musi Charitas Catholic University. The data used in this research consist of academic records of students from 2018 to 2020, including Grade Point Average per semester (GPA) and Cumulative Grade Point Average (CGPA). The research method follows the Knowledge Discovery in Database (KDD) stages, namely data selection, preprocessing, transformation, data mining, and interpretation. Model development and evaluation were conducted using RapidMiner software with the 10-Fold Cross Validation method. The results indicate that Semester 8 GPA is the most influential attribute in determining student graduation status, followed by Semester 4 GPA as a supporting indicator. The generated decision tree model achieved an accuracy rate of 75.68%, indicating a good predictive performance. These findings demonstrate that the C4.5 Decision Tree algorithm can serve as an effective decision-support tool for early detection of students at risk of delayed graduation, thereby assisting academic institutions in improving on-time graduation rates and academic management quality.]]>
Copyrights © 2025
