Implementasi Algoritma K-Means dan C-Means untuk Clustering Angka Kemiskinan

Abstract

Implementation of K-Means and C-Means Algorithms for Clustering Poverty Numbers - Poverty is one of the problems that must be faced by developing countries, including Indonesia and especially the province of West Java. This problem is exacerbated by the Covid-19 pandemic. Poverty can also have other consequences, such as increased crime and death. To facilitate government programs and support, it is necessary to group cities/districts according to the poverty level. The analysis was carried out using the K-Means and Fuzzy C-Means algorithms with the Silhouette method to obtain the optimal number of clusters using RStudio tools. The purpose of this study is to compare which algorithm is based on the Davis-Bouldin Index validation test. Three of the five data generated, the K-Means and C-Means algorithms give the same results. Only poverty data and education data give different results. Based on the results of the Davies-Bouldin Index validation test, the fuzzy c-means and k-means algorithms show that the k-means algorithm is better at clustering with an average of 4.084271. Meanwhile, fuzzy c-means has an average validation score of 4.111375. The smaller the Davies-Bouldin Index value or the closer to 0 shows how good the cluster is.