Skripsi
KLASTERISASI WILAYAH DI INDONESIA BERDASARKAN INDIKATOR PREVALENSI KETIDAKCUKUPAN KONSUMSI PANGAN MENGGUNAKAN K-MEANS CLUSTERING DAN UJI PERBEDAAN ANTAR KLASTER
The Prevalence of Undernourishment (PoU) is an important indicator for describing the state of food security in Indonesia. This study aims to Cluster regency/city in Indonesia based on seven PoU-related indicators. The Clustering method used is K-Means Clustering. The indicators analyzed include Rice Production, Population Density, Percentage of Poor Population, Percentage of Per Capita Expenditure on Food, Desirable Dietary Pattern, Human Development Index, and Food Security Index. The data consist of 514 regency/city in Indonesia obtained from Badan Pusat Statistik (BPS) and Badan Pangan Nasional (Bapanas) for the year 2024. The determination of the number of clusters in the K-Means method was carried out by testing 2, 3, 4, and 5 clusters, which were evaluated using the Silhouette Coefficient. The highest Silhouette value was obtained for three clusters, with a value of 0.3444. Based on this result, the regencies/cities were grouped into three clusters, consisting of 84 regencies/cities in Cluster 1, 349 in Cluster 2, and 81 in Cluster 3. Cluster 1 shows relatively high values for the percentage of poor population and per capita food expenditure. Cluster 2 exhibits relatively high values for rice production, per capita food expenditure, Desirable Dietary Pattern, and Food Security Index. Cluster 3 has relatively high values for population density, Human Development Index, Desirable Dietary Pattern, and Food Security Index. The comparison of mean PoU values across the three clusters using the Kruskal–Wallis test indicates a significant difference in PoU values among clusters, suggesting that each cluster has distinct PoU characteristics based on the forming indicators. Keywords : Cluster, PoU, K-Means Clustering, Silhouette, Kruskal Wallis
No other version available