Penerapan Metode K-Harmonic Means pada Pengelompokan Kabupaten/Kota di Pulau Kalimantan Berdasarkan Indikator Perumahan dan Kesehatan Lingkungan

Abstract

Cluster analysis is grouping data into subsets whose members have a high degree of similarity. One commonly used clustering technique is the K-Harmonic Means, an enhancement of the K-Means method, which assigns data to clusters by calculating the harmonic mean of distances from each point to the cluster centers. This study aims to determine the optimal grouping of Regencies/Cities in Kalimantan Island based on housing and environmental health indicators and to identify the best Dunn index value. Based on the grouping results, 5 groups were obtained with parameters (p) = 2. The optimal Dunn index value is 0.2171. Based on the description of the housing and environmental health indicators, it can be seen that there are still disparities between groups in terms of access to basic services, such as the percentage of clean drinking water source services, the percentage of lighting sources from PLN electricity, the percentage of proper sanitation services, and the percentage of home ownership status. Group 4 and group 2 are groups that need special attention because they have the lowest achievements in several important indicators