Pengelompokan Data Stunting di Indonesia Menggunakan Metode X-Means dan Agglomerative Hierarchical Clustering

Stunting is one of the serious problems that threaten the quality of human resources in Indonesia. This study aims to analyze the patterns and characteristics of stunting in Indonesia by applying the X-Means clustering method and Agglomerative Hierarchical Clustering (AHC). The X-Means method is used to determine the optimal number of clusters automatically by utilizing the Bayesian Information Criterion (BIC), while AHC forms a dendrogram to understand the multilevel structure of the clusters formed. Based on the analysis, the X-Means method produces three optimal clusters with the smallest BIC value of 651.9475, where cluster 1 consists of 17 provinces, cluster 2 includes 12 provinces, and cluster 3 includes 5 provinces. The AHC method with the Single Linkage approach also produced three optimal clusters, with cluster 1 covering 32 provinces, cluster 2 consisting of 1 province (West Nusa Tenggara), and cluster 3 covering 1 province (East Nusa Tenggara), as well as the highest Silhouette Index value of 0.28. The results show that both methods provide a comprehensive picture of stunting patterns in Indonesia, which can be used as a basis for designing more targeted intervention programs according to the characteristics of each cluster. This data-driven strategy is expected to increase policy effectiveness in reducing stunting in Indonesia.