Analisis Cluster Fuzzy C-Means dan Diskriminan untuk Pengelompokan Data Kesejahteraan Rakyat

Cluster analysis is a multivariate method that aims to classify objects into a cluster based on different characteristics. The statistical technique that can be used for clustering is Fuzzy C-means (FCM). The FCM uses membership degree to determine the existence of each data point in a cluster. To ensure accurate clustering and that the criteria are met, cluster validation must be done to produce good data clustering. Moreover, this study uses a discriminant analysis test to validate the results of cluster solutions. The study aims to classify 34 provinces in Indonesia based on the level of people's welfare. The data used in this study is data on indicators of people's welfare in Indonesia in 2021. The variables used in this study were age/life expectancy, the average length of schooling, the 16-18 years of enrollment, and households with proper access to water. Based on the study results, two clusters were formed: cluster 1, areas with low people’s welfare indicator values in 16 provinces, and Cluster 2, regions with high people's welfare indicator values in 18 provinces.