Comparison of K-Means and Fuzzy C-Means Algorithms for Clustering Based on Happiness Index Components Across Provinces in Indonesia

Abstract

Cluster analysis is a statistical technique used to group objects based on their shared characteristics. This research aims to assess how 34 provinces in Indonesia are clustered using happiness index indicators for the year 2021. The study compares two non-hierarchical cluster analysis methods, K-Means and Fuzzy C-Means. K-Means categorizes objects into clusters based on their proximity to the nearest cluster center, while Fuzzy C-Means employs a fuzzy grouping model assigning membership degrees from 0 to 1. The results indicate that both methods form three clusters. Evaluating standard deviation values and ratios, Fuzzy C-Means proves superior, displaying a larger standard deviation between groups and a smaller ratio (0.6680004) compared to K-Means. Consequently, the study concludes that the Fuzzy C-Means method is more optimal than K-Means.