Stunting, a condition where children are shorter than their age, is a serious nutritional issue in developing countries, including Indonesia. Research shows that Low Birth Weight (LBW) significantly affects children’s growth. In Bekasi Regency, the stunting prevalence remains high at 17%, with a target reduction to 14%. Achieving this target requires prevention efforts focused on improving nutrition and regularly monitoring child growth. In such monitoring, innovative eval uation techniques using Machine Learning (ML) are needed to predict stunting potential. This study aims to develop a predictive model for early detection of stunting risks in Bekasi Regency, using machine learning techniques to analyze Low Birth Weight (LBW) and Low Birth Length (LBL) factors, which could potentially be integrated into the local health moni toring system for preventive intervention. RF excels in handling complex features and identifying important predictors, while KNN is effective at recognizing local patterns. The results show that RF achieved the best performance with 99.22% accuracy and an F1-score of 96.94%, compared to KNN with 96.19% accuracy and an F1-score of 87.16%, highlighting RF’s greater stability and robustness over KNN in predicting stunting cases. This study is expected to provide an accurate predictive system that helps parents, health workers, and the government identify stunting potential early while also determining the appropriate ML algorithm for stunting case prediction in Indonesia. Future research is encouraged to test this model in other regions with different characteristics to ensure the generalizability and effectiveness of stunting prediction on a broader scale.