Implementation of Data Mining for Nutrition Clustering of Toddlers at Posyandu Using K-Means Algorithm

This study aims to determine the nutritional status of toddlers in the Peulimbang sub-district using a clustering method that can group toddlers based on their dietary indicators, such as gender (jk), age (u), height (tb), weight (bb), and upper arm circumference (Lila). By using data analysis techniques such as K-Means clustering, this study successfully identified several groups of toddlers with different nutritional statuses, ranging from malnutrition to good nutrition to obesity nutritional status. The data for this study were taken directly from the Peulimbang Health Center, and as many as 765 toddler data were collected from 16 villages in the Peulimbang area. The programming language used in this study is PHP, which functions in web development and is often used to process data sent via web formula, interact with databases, and manage user sessions. Based on the results of clustering with the K-Means method using Euclidean Distance as a measurement between points, toddler data has been grouped into three Clusters, namely C1, C2, and C3. Cluster C1 covers 22.81% of 147 malnourished toddlers, C2 covers 48% of 323 well-nourished toddlers, and C3 covers 29.19% of 205 obese toddlers. Based on the clustering results, improving nutrition education programs for parents is essential, especially in areas with poor or lacking nutritional status. This program can include counseling on the importance of balanced nutrition, nutritious cooking methods, and choosing the right Food for toddlers. Thus, this study is expected to contribute to improving toddlers' nutritional status and overall public health in the Peulimbang District area.