Analisis Faktor Risiko Stunting untuk Perbaikan Gizi di Indonesia Menggunakan Model Machine Learning

Stunting is a significant nutritional issue in Indonesia, impacting the physical and cognitive development of children. This study aims to identify the risk factors contributing to stunting and design more targeted nutrition improvement policies through data analysis. Data was obtained from the Kaggle platform, which includes various demographic, economic, and health factors influencing stunting. The machine learning models applied in this study are Random Forest, Decision Tree, and Naive Bayes. These models were selected based on their proven effectiveness in previous research on health and nutrition data analysis. The results of this study show that Random Forest provides the best accuracy and greater stability, while Decision Tree offers clearer interpretations of the risk factors for stunting. Naive Bayes, though simpler, still provides useful results as a baseline model. By using k-fold cross-validation, this study ensures that the models can generate accurate and robust predictions. The findings are expected to provide a solid foundation for evidence-based nutrition policies in efforts to improve nutrition and reduce stunting in Indonesia.