<![CDATA[Pulmonary tuberculosis (TB) remains a major public health concern, particularly in Indonesia, where the burden of the disease remains high. Although the Directly Observed Treatment Shortcourse (DOTS) strategy has been implemented to control TB, the detection of new cases remains a challenge. This study aims to develop a pulmonary TB risk prediction model using the CatBoost algorithm based on patient data from Kabangka Community Health Center, Muna Regency. A total of 2,724 patient records were used in the analysis. The research process involved data preprocessing, dataset splitting, model training, and performance evaluation using Mean Absolute Error (MAE), Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE). The evaluation results yielded an MAE of 0.00467, an MSE of 0.00385, a MAPE of 36.37%, and an RMSE of 0.062. These results indicate the prediction errors obtained from the implemented model and provide an initial assessment of its performance on the dataset. The model is expected to support preliminary analysis of pulmonary TB risk and provide information that may assist public health planning and interventions.]]>
Copyrights © 2026
