Stunting, a serious public health concern in Indonesia affects millions of children globally and is caused by chronic malnutrition in early life. This disorder, which affects over 165 million children under five worldwide, is the most common kind of undernutrition and is defined by a height that is much below normal for a child's age. The diagnosis of stunting is made when the height-for-age Z-score (HAZ) is less than -2. This work uses a dataset of more than 8,000 records from the Karya Jaya Health Center to create and assess machine learning models for toddler stunting prediction. Two machine learning algorithms, K-Nearest Neighbors (KNN) and Support Vector Machines (SVM), were analyzed for their predictive performance. The Synthetic Minority Oversampling Technique (SMOTE) and outlier treatment are two crucial preprocessing methods that improved model dependability and corrected data imbalances. In terms of total performance metrics and an amazing accuracy of 99.2%, the results show that KNN performed better than SVM. These results demonstrate the potential of machine learning to generate precise stunting projections in order to address this pervasive issue allowing for prompt treatments and focused public health initiatives.