Klasterisasi Data Penggunaan Layanan BPJS Kesehatan Menggunakan Algoritma K-Means

Abstract

BPJS Health, which is organized by the government by upholding the principle of mutual assistance in equalizing public health insurance, many patients use these facilities. The clustering method is processed with the K-Means algorithm, where the results also show a new insight, namely grouping the use of BPJS health services based on 3 clusters. Cluster 1 is a category of health facilities with low or Low use of BPJS health services, which is 353 out of 1000 categories of health facilities based on the number of BPJS health service usages tested, then cluster 2 is a category of health facilities with moderate or Medium use of BPJS health services, which is 474 out of 1000 categories. the name of the health facility based on the number of health BPJS service usage tested, and lastly cluster 3 is the category of health facility name with the use of BPJS health services quite high or High, namely 173 out of 1000 categories of health facility names based on the number of health BPJS service usage tested. Tests using Rapid Miner tools can also produce similar insights, namely each cluster has cluster group members according to manual calculations such as Cluster_0 on Rapid Miner has 474 cluster members representing the Medium cluster, Cluster_1 has 353 cluster group members as the Low cluster representation, and Cluster_2 has 173 cluster members that correspond to the cluster representation of High.

Keywords: Data Mining, K-Means, BPJS Health