<![CDATA[Abstract—Nutritional status in early childhood is a key indicator of public health and serves as a basis for targeted nutritional interventions. This study applies the K-Means Clustering algorithm to anthropometric data—including weight, height, and head circumference—to classify children's nutritional status into five categories: severely undernourished (−3 SD), moderately undernourished (−2 SD), normal, mildly overnourished (+1 SD), and moderately overnourished (+2 SD). The clustering process is based on Z-scores derived from WHO standards, namely Weight-for-Age (WAZ), Height-for-Age (HAZ), and Head Circumference-for-Age (HCAZ), which serve as input features for the clustering algorithm. The number of clusters (k = 5) is aligned with national nutritional classification guidelines. The clustering results are visualized to illustrate the data distribution across the three indicators, and evaluated using Euclidean distance to assess the proximity of each data point to its assigned cluster centroid. The results demonstrate that K-Means Clustering can effectively classify children's nutritional status in a manner consistent with manual classification based on WHO thresholds. This study highlights the potential of data mining approaches in supporting health information systems for the early and automated detection of child malnutrition. Index Terms—Nutritional Status; Anthropometry; Z-Score; K-Means Clustering]]>
Copyrights © 2026
