<![CDATA[Background:As the database grows, predicting students' academic performance becomes more difficult. Traditional methods often overlook students with exceptional achievements and fail to fully track their progress. Although traditional assessments like exams and assignments provide valuable insights, they may not consider all factors affecting performance, such as socioeconomic status and engagement rates. Aims: This study develops a predictive model aimed at classifying students' academic performance in higher education. Methods: Using a combination of machine learning algorithms. Data collected from the Department of Computer Science and the Department of Mathematics at Tai Solarin University of Education was analyzed through the mutual information method to identify important factors. The model was created and tested using Google CoLaboratory, employing two algorithms: Support Vector Machines (SVM) and Decision Trees (DT). The accuracy of the models was measured using important indicators, including accuracy, precision, and the F-measure. Results:This study shows that machine learning techniques can effectively identify student performance early, with SVM achieving 100% accuracy, enabling quicker involvements and better resource allocation. Conclusion: Additionally, it supports evidence-based decision-making in educational institutions, which helps improve student encounter and enhances learners retention.]]>
Copyrights © 2026
