<![CDATA[Food security is a vital aspect of national development in Indonesia, given its diverse geographical and socioeconomic conditions. This study aims to cluster 514 regencies/cities in Indonesia based on food security indicators and compare clustering methods to obtain optimal results. Secondary data from 2024 from the Central Bureau of Statistics and the National Food Agency were analyzed, covering variables such as poverty, electricity access, female mean years of schooling, food expenditure, and health worker ratios. The evaluated methods were K-Means and Ward, assessed using the Silhouette Score, Davies-Bouldin Index, Dunn Index, Separation Index, and Sw/Sb Ratio. Results show that K-Means achieved a higher Silhouette Score, whereas Ward excelled with a lower Davies-Bouldin Index, higher Dunn Index, and a cophenetic correlation coefficient of 0.997. Consequently, the Ward method was selected as the superior approach, forming four distinct clusters: Cluster 1 (highly vulnerable), Cluster 2 (highly secure), Cluster 3 (secure), and Cluster 4 (moderately secure). The clustering highlights significant regional food security disparities, with urban areas generally demonstrating better food security than underdeveloped regions.]]>
Copyrights © 2026
