Pengelompokan Kabupaten/Kota di Kalimantan Berdasarkan Indikator Pendidikan Menggunakan Metode K-Means dengan Optimasi Principal Component Analysis

Abstract

Cluster analysis is used to group several objects based on similarities within the group. There are many methods included in cluster analysis, including k-means. K-means is a non-hierarchical cluster analysis method. The assumption that needs to be considered in cluster analysis is that there is no strong correlation between research variables. An alternative that can be done to deal with variables that are strongly correlated is to use Principal Component Analysis (PCA). This research aims to group districts/cities in Kalimantan based on education indicators in 2022 using k-means with PCA optimization, as well as finding out the optimal cluster based on the smallest Davies Bouldin Index (DBI) value. Based on the results of the analysis, from 11 research variables two main components were formed. From these two main components, new data transformations are produced which are then used in grouping districts/cities in Kalimantan based on education indicators using the k-means methods. The analysis results, it was found that the optimal cluster with k-means grouping was 5 clusters with a DBI value of 0.835. Cluster 1 has 8 regencies/cities, cluster 2 has 16 regencies/cities, cluster 6 has 5 regencies/cities, cluster 4 has 21 regencies/cities, and cluster 5 has 5 regencies/cities.