Advanced Machine Learning Techniques for Stunting Classification Using a Stacking Ensemble Approach

Stunting is a critical public health issue that severely impacts children's physical and cognitive development, making it a prominent concern within the Sustainable Development Goals. This study focuses on classifying stunting cases in Indonesia using an advanced stacking ensemble method, integrating Decision Tree, Random Forest, K-Nearest Neighbors (KNN), XGBoost, and Gradient Boosting algorithms. The analysis utilizes primary data from the 2023 Indonesian Health Survey conducted by the Ministry of Health, comprising 86,364 observations and 19 variables related to social, economic, environmental, and demographic factors. The study compares the performance of individual algorithms with a stacking ensemble approach, which combines the strengths of multiple algorithms. The findings reveal that the stacking ensemble approach, particularly when combining XGBoost and Gradient Boosting, achieved the highest accuracy of 0.79, outperforming individual algorithms like the Decision Tree, which had the lowest accuracy of 0.66. These results highlight the superior effectiveness of the stacking ensemble method, especially the XGBoost and Gradient Boosting combination, in accurately classifying stunting cases, providing valuable insights for targeted interventions and public health strategies.