<![CDATA[The increasing number of HIV cases in Indonesia, particularly in West Java, which ranks among the regions with the highest infection rates, highlights the need for forecasting models capable of producing accurate estimates to support health policy planning. This situation underscores the importance of analytical approaches that can capture the dynamic progression of cases over time. This study aims to develop a forecasting model for the number of HIV cases in West Java by applying the CRISP-DM framework and utilizing a dataset categorized by age group for the period 2019–2023. Two regression algorithms, Support Vector Regression (SVR) and Decision Tree Regression (DTR), were compared using three evaluation metrics: Coefficient of Determination (R²), Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE). The findings indicate that DTR delivers superior performance, achieving an R² of 83.62, RMSE of 32.29, and MAPE of 33.30. In contrast, SVR produced an R² of 36.33, RMSE of 84.01, and MAPE of 44.87. Based on these results, Decision Tree Regression is identified as the more effective model for forecasting the number of HIV cases in West Java, providing practical support for health policy decision-making, resource allocation, and prioritization of targeted prevention and intervention programs at the regional level.]]>
Copyrights © 2026
