<![CDATA[Student dropout continues to pose a critical challenge that hinders educational equality and long term human capital development. This study aims to predict dropout risk by employing attendance records and parental occupation as the main indicators. A synthetic dataset reflecting Indonesian school conditions was analyzed using the Decision Tree algorithm. The results demonstrate that the model achieved strong predictive capability, reaching 87% accuracy with well balanced precision and recall values. Further analysis highlights absenteeism as the most decisive factor influencing dropout risk. In addition, parental occupation emerges as a contextual determinant that strengthens risk identification, with students whose parents are engaged in informal or unstable sectors being more vulnerable compared to peers from households with stable formal employment. The transparent structure of the Decision Tree enhances its practical value for educational practitioners, as it translates complex data into insights that are both actionable and accessible. While the findings are based on simulated data, the study underscores the importance of integrating behavioral and socioeconomic indicators into early detection frameworks for student dropout.]]>
Copyrights © 2025
