K-Means Clustering Analysis of Poverty Data in Cilacap District

Abstract

Poverty stands as a complex structural obstacle within social development frameworks. The COVID-19 pandemic intensified poverty dynamics in Indonesia which saw poverty rates increase by 9.78% in March and reach 10.19% by September. Local Bureau of Statistics data shows that the poverty rate in Cilacap Regency dropped to 10.99% (around 191,000 people) in March 2024 from 10.68% (186,080 people) in March 2023. The study uses k-means clustering methodology for analysis and maps poverty-prone areas utilizing QGIS software. The analysis revealed 12 sub-districts and 14 neighborhood units (RW) alongside a single community unit (RT) that show unique poverty characteristics. The silhouette coefficient evaluation produced a 0.55 score which showed a moderate cluster structure and acceptable cluster placement. The research provides empirical evidence about poverty distribution which shows how data mining methods can enhance spatial socioeconomic studies. The study presents a detailed analysis of poverty stratification across Cilacap Regency through the application of sophisticated computational methods.