<![CDATA[On-time student graduation is a key indicator of academic effectiveness and higher education quality. Timely graduation reflects efficient academic management, while delays may negatively impact institutional performance and accreditation. Graduation delays are influenced by various academic and non-academic factors, making early prediction essential. This study aims to develop an on-time graduation prediction model using the C4.5 algorithm optimized with the forward selection method. The research was conducted in the Computer Engineering Study Program, Faculty of Computer Science, Universitas Brawijaya, using student data from the 2018–2021 cohorts. The dataset includes both academic and nonacademic attributes. The modeling process followed the CRISP-DM framework, and model performance was evaluated using a confusion matrix. The results show that the C4.5 model without feature selection achieved an accuracy of 63%, while the application of forward selection significantly improved accuracy to 83%. These findings indicate that feature selection plays a crucial role in enhancing prediction performance. The proposed model can support academic stakeholders in data driven decision making and in designing strategies to improve on-time graduation rates. ]]>
Copyrights © 2026
