Penerapan Data mining untuk Clustering Kondisi Sosial Ekonomi Berdasarkan Kepemilikan Jaminan Kesehatan Menggunakan Algoritma K-Means

Abstract

This study discusses the application of the K-Means Clustering algorithm in analyzing the socioeconomic conditions of communities in North Sumatra Province based on health insurance coverage. The background of this study stems from the continuing gap in access to and coverage of health insurance, which is influenced by differences in socioeconomic conditions between regions. The purpose of this study is to identify community groups with different patterns of health insurance ownership, analyze their influence on socioeconomic conditions, and explain the application of the K-Means algorithm as a data clustering method. The data used was obtained from the North Sumatra Provincial Statistics Agency (BPS) for the 2021–2023 period, with variables including BPJS Health Insurance Premium Assistance Recipients, BPJS Health Insurance Non-Premium Assistance Recipients, Private Insurance, and Insurance from Companies/Offices. The analysis process was carried out through the stages of variable selection, initial centroid determination, distance calculation using Euclidean Distance, and iteration. The results of the study show that there are two main clusters, namely clusters with high health insurance ownership rates and clusters with low ownership rates. In terms of the BPJS Health Insurance Variable for Contribution Assistance Recipients, the cities of Medan, Samosir, Tanjungbalai, Asahan, Batu Bara, Binjai, and Toba are included in the high cluster, while Deli Serdang is in the low cluster. These findings are expected to provide an overview for local governments in formulating policies to improve access to health insurance in a more targeted and data-driven manner.