<![CDATA[Cluster analysis is a multivariate analysis aimed at grouping objects based on specific shared characteristics of the objects under study. This research aims to determine the clustering results of regencies/cities in Java Island using the k-medoids and Clustering Large Applications (CLARA) methods with three distance measures: Euclidean, Manhattan, and Minkowski, based on the Human Development Index (HDI) indicators. The data used in this study is the 2023 HDI indicator data consisting of 4 variables and 119 regencies/cities. Based on the analysis, the combination of the CLARA method and Euclidean distance has the highest silhouette coefficient value of 0.380, resulting in clusters with members of 39, 26, 43, and 11 regencies/cities respectively. This indicates that, in this case, the CLARA method is better suited for use, aligning with the initial theory that this algorithm is more appropriate for data with a minimum of 100 observations. Governments in the regencies/cities of Java Island can use the clustering results to help design more targeted and effective development programs for each region.]]>
Copyrights © 2025
