Analysis of the Grouping of Provinces in Indonesia According to the Democracy Index With the Agglomerative Hierarchical Clustering Algorithm

Abstract

The purpose of this study is to clustering the provinces in Indonesia based on the IDI dimensions in 2020. This study compares four methods of agglomerative hierarchical clustering, namely the single linkage method, average linkage, complete linkage, and the ward method. The evaluation criteria used are the silhouette index, dunn index, connectivity index, and agglomerative coefficient. The best method used is the average linkage hierarchy method with two clusters. The combination of two clusters and the ward’s method produces a silhouette index value of 0.34, a dunn index value of 0.28, a connectivity index value of 5.64 and an agglomerative coefficient of 0.891 which illustrates that the method is able to classify provinces in Indonesia based on the 2020 IDI. The results of this study can be used as a consideration for the government to take policies related to policies in Indonesia.