<![CDATA[Equitable funding for school infrastructure is often hampered by subjective evaluation methods that lead to inappropriate prioritization. This study develops a data-driven approach that uses machine learning to objectively classify funding priority areas. Using the CRISP-DM (Cross-Industry Standard Process for Data Mining) framework and data from the Indonesian Ministry of Education, a support vector regression (SVR) model with an RBF kernel is developed. This model integrates key facility indicators including laboratory availability, sanitation, and access to utilities to predict infrastructure adequacy. Separate models were built for public and private secondary schools with a 70:15:15 data split. The results demonstrated excellent predictive accuracy, with an R² of 0.9938 for public schools and 0.9969 for private schools, at a minimal error rate (MAE <0.20). By grouping the regression results into priority categories, the model successfully identified twenty high-priority areas that require immediate intervention by 2024. These results demonstrate that the SVR-based framework provides a robust decision support system, enabling policymakers to allocate infrastructure funds more transparently, equitably, and in direct alignment with empirical realities on the ground.]]>
Copyrights © 2026
