Klasifikasi Berisiko Stunting pada Balita: Perbandingan K-Nearest Neighbor, Naive Bayes, Support Vector Machine

Stunting in children under five is a significant health problem that impacts child development. This study aims to develop a classification model to predict stunting risk using SVM, KNN, and Naïve Bayes algorithms. Data from the Jatilawang Health Center included 523 under-fives with variables such as age, weight, length, arm circumference, z-score, parental education, and maternal health history. Following the CRISP-DM steps, the data was processed through handling missing data, feature selection, and dividing the data into training and testing sets with a ratio of 80:20. Results showed SVM had the highest accuracy of 90%, followed by KNN 89%, and Naïve Bayes 85%. This research produces a stunting risk prediction model that is implemented in a simple website, supporting early intervention and decision-making in stunting prevention efforts.