Application of the K-Means Clustering Method to Cluster Stunting Cases Based on Family Economics in Langkat Regency

Abstract

Stunting in children is a serious health issue that has long-term impacts on the quality of human resources in Indonesia. Langkat Regency is one of the regions with a high prevalence of stunting. Family economic factors, such as parents' occupation and housing conditions, are suspected to play a significant role in influencing children's nutritional status. However, there is still a lack of data-based studies that specifically cluster stunting cases based on these factors. To address this need, this study applies the K-Means Clustering method to group stunted children based on three main variables: parents' occupation, housing status, and causes of stunting. This algorithm was chosen for its effectiveness in identifying hidden patterns within medium-sized data. The clustering process involved data transformation, determining the number of clusters, calculating distances using Euclidean Distance, and iterative processing to obtain the optimal centroid. The implementation was carried out using MATLAB R2014b software with stunting data obtained from the PPKB-PPA Office of Langkat Regency for the years 2023–2024. The results of the study yielded three main clusters representing the family's economic condition and its relationship to stunting. The patterns found indicate that children from families with unstable jobs and inadequate housing tend to be more vulnerable to stunting. These findings provide a strong foundation for the formulation of more targeted policies in addressing stunting by local governments.