<![CDATA[Stunting is a chronic nutritional problem that has long-term effects on children's physical growth and cognitive development. Therefore, classifying the nutritional status of toddlers is an important step in early detection and determining appropriate interventions. This study aims to compare the performance of two classification methods, namely Gaussian Naïve Bayes and K-Nearest Neighbor (K-NN), in identifying the nutritional status of toddlers. The data used consisted of 14,620 toddler data obtained from the Takalar District Health Office covering 11 sub-districts. Gaussian Naïve Bayes is a probabilistic classification method with the assumption of independence between variables, while K-NN is a nonparametric method that classifies data based on the proximity of the distance between observations. The results showed that Gaussian Naïve Bayes produced an accuracy of 91.76%, but was unable to accurately classify stunting classes due to class imbalance and low posterior probability values in minority classes. In contrast, the K- NN method with an optimal parameter value of k=3 produced an accuracy of 97.00% and showed better performance in identifying toddlers with stunting status. Based on these results, the K-NN method is considered superior to Gaussian Naïve Bayes in classifying the nutritional status of toddlers in Takalar Regency.]]>
Copyrights © 2026
