Application of data mining techniques using the K-Means Method on Unmet Need of Health Services by Province in Indonesia

Abstract

Abstract The research objective is to cluster the Unmet Need of health services in Indonesia by region by utilizing data mining techniques. The data source was obtained from the Central Statistics Agency (abbreviated BPS) in the health category using Unmet Need data on health services by province (2015-2018) consisting of 34 data records. The variable used is the Unmet Need of health services by province (percent). The data mining technique used is k-means which is part of clustering. Data processing is performed using RapidMiner 5.3 software. The cluster labels used in the study are high cluster (C1) and low cluster (C2). The results state that the application of k-means can be carried out where 9 provinces are in cluster C1 consisting of Riau, West Nusa Tenggara, East Nusa Tenggara, Central Kalimantan, South Borneo, Central Sulawesi, Southeast Sulawesi, Gorontalo and West Sulawesi with final centroids (C ) = 6.96 and 25 provinces are in cluster C2 with final centroid (C) = 4.36. The results of mapping in the form of regional clusters can provide information for the government and should Unmet Need health services in Indonesia must be worth 0% for each region.