A Spatial Approach to Analyze the Distribution and Risk Factors of Stunting in North Sumatra With the K-Means Algorithm

This research aims to analyze the spatial pattern of stunting incidents in North Sumatra based on family economic factors using the K-Means algorithm. The data used includes the incidence of stunting, family economics, and spatial data (geographical coordinates) from the Central Statistics Agency (BPS) and Basic Health Research (Riskesdas). The elbow and silhouette analysis methods determine the optimal number of clusters, resulting in four clusters: Cluster 1 with varying economic conditions but very limited access to health and sanitation and a high incidence of stunting; Cluster 2 with low economic conditions, limited access to health, and a high incidence of stunting; Cluster 3 with medium economic conditions and moderate stunting incidence; Cluster 4 with good economic conditions and health services and a low incidence of stunting. Validation using the Silhouette Coefficient shows an average value of 0.553, indicating good clustering quality. The analysis shows that family economic factors, access to health services, and sanitation conditions have a significant effect on the incidence of stunting. Policy and intervention recommendations are focused on clusters 1 and 2 with interventions including increasing access and quality of nutrition, health services, improving sanitation, economic empowerment, and health education and counseling.