Penerapan Metodetwo Step Cluster Dalam Analisis Gerombol (Studi Kasus : Data Potensi Desa Sensus Ekonomi 2003 Wilayah Jawa Barat)

Abstract

Clustering is a process of grouping objects based on similarities . Some problems in cluster analysis are the different scale of measurement, large data sets and unspecified number of cluster. One approach to handle this problems is by using Two Step cluster method. The purpose of this research is to applicate Two Step cluster method by clustering villages residing in West Java. The result of Two Step cluster method are pre-cluster which early yielded at first phase counted eight clusters and optimal cluster which yielded at second phase counted three clusters. Members in first cluster are objects which is extreme outliers and cannot be grouped into other clusters. Villages which include into second clusters have rural characteristic. The village has wide farm, many agriculture households, but also has underdeveloped field of industries and also information and communications. So that, to improve village potency for this cluster, the variables which still have low growth level has to be increased. The third cluster has the characteristic of urban status. Village in this cluster has the closest distance to downtown, developed enough in the field of industry, information and communications, but still have the highest number for unemployment citizens.