OPTIMALISASI CLUSTER PADA CLUSTER HIERARKI MENGGUNAKAN PSEUDO F-STATISTIC CALINSKI HARABASZ UNTUK KETAHANAN PANGAN

Abstract

Food security is the ability of a nation to ensure that all of its inhabitants obtain food in sufficient quantities, of proper quality, based on optimizing the utilization of food. The food security of each region is different, resulting a region to have low or high food security, one of which is the area in Pati Regency as a research area. To find out this, area clustering was carried out based on variables of planted area, food production, number of food supply facilities and infrastructure, number of people with low welfare levels, number of sub-districts without sufficient connecting access, number of households without access to clean water, area, number of residents, number of households, and level of settlement. The grouping of food security areas was carried out using the Hierarchical Clustering method based on Euclidean distance calculations with cluster validation using Calinski Harabasz Pseudo F-Statistics. The results of this study obtained the optimum number of clusters as 7 clusters in the Complete Linkage closest distance calculation technique which has the highest Pseudo-F value. High quality food security is found in the highlands, namely, Gembong District, Cluwak District, Trangkil District, and Gunung Wungkal District. Meanwhile, the low quality of food security is found in lowland areas, namely, Juwana District and Pati District.