<![CDATA[Student dropout is a critical issue in higher education. This study aims to develop a predictive model of dropout based on early academic performance using Random Forest and Gradient Boosting algorithms. The dataset, sourced from the UCI Repository, contains 4,424 student records. Key features analyzed include the number of enrolled courses, evaluations, average grades, and enrollment age. Results show that the Gradient Boosting algorithm achieved 70.05% accuracy, while Random Forest reached 70.16%, both performing best in classifying graduates. The model successfully identifies high-risk students, although challenges remain in predicting “enrolled” status. These findings highlight the potential of machine learning for early dropout detection and support more targeted academic interventions.]]>
Copyrights © 2026
