ARCHIVES

Student dropout is a major challenge in higher education, often linked to academic, demographic, and socioeconomic factors. This study uses machine learning techniques to predict student dropout and support early intervention. A dataset containing attributes like attendance, CGPA, internet access, parental education, and socioeconomic status was analyzed. Data preprocessing, feature selection, and model evaluation were performed using algorithms such as Random Forest, Logistic Regression, and XGBoost. Among these, XGBoost achieved the highest accuracy of 87%. The findings show that data-driven models can effectively identify at-risk students, helping institutions make informed decisions to improve retention