<![CDATA[Income inequality among provinces in Indonesia reflects differences in welfare levels influenced by the social and economic characteristics of each region. This study aims to compare the performance of Par-ticle Swarm Optimization (PSO) and Firefly Algorithm (FA) in optimizing the number of clusters in K-Means clustering to group Indonesian provinces based on seven factors related to income inequality, namely Human Development Index (HDI), Number of Poor Population, Open Unemployment Rate, In-flation, Provincial Minimum Wage, GDP per Capita, and Labor Force Participation Rate. The data used are secondary data from 2024 covering 38 provinces. The analytical methods include data standardi-zation using Z-Score, K-Means clustering, and cluster number optimization using PSO and FA evalu-ated by Silhouette Coefficient (SC). The results show that both optimization methods improved clus-tering quality compared to K-Means without optimization, which yielded an SC of 0.23. PSO produced an SC of 0.28 with an optimal cluster number of 5, while FA produced an SC of 0.39 with an optimal cluster number of 3. FA proved superior in generating a more optimal and representative clustering structure. The clustering results reveal distinct characteristics among clusters that can serve as a ba-sis for formulating more targeted income inequality reduction policies in accordance with the charac-teristics of each regional group.]]>
Copyrights © 2026
