<![CDATA[This study addressed the need for an effective approach to predicting student academic performance in higher education using data-driven methods. The study aimed to implement machine learning models to predict the academic performance of students in the Information and Communication Technology Education Study Program at Universitas Negeri Medan. A quantitative predictive design was employed using a dataset of 40 student records. Five classification models were tested, namely Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and Naïve Bayes. The results showed that all models produced strong predictive performance. Decision Tree achieved the highest accuracy at 93.1%, Logistic Regression produced the highest precision at 95.9% and the highest F1-score at 93.2%, while Support Vector Machine obtained the highest recall at 93.2%. These findings indicated that machine learning was feasible for predicting student academic performance in the study program. The study concluded that Logistic Regression provided the most balanced overall performance and had strong potential to support early academic intervention and data-based academic decision making in higher education.]]>
Copyrights © 2026
