Data Mining K-Means: Clustering Health and Complaints Resident in Indonesia

Abstract

This study aims to utilize the Clustering Algorithm in grouping the population Which have complaint health with algorithm K-means in Indonesia. Source data study This collected based on the information documents. The total population of the province have complaints health produced by the Central Bureau of National Statistics. The data used in this study are data for 2013-2017 which consists of 34 provinces. The method used in this research is K-means algorithm. The data will be processed by clustering in 3 clusters, namely level clusters high health complaints, clusters of moderate and low health complaints. Data center for clusters high population level 37.48, Centroid data for clusters of moderate population level 27.08, and Centroid data for low population level cluster 14.89. So that the acquisition of the assessment is based on the population index owned health complaints with 7 provinces with high levels of health complaints, namely Central Java, in Yogyakarta, Bali, Nusa Southeast West, Nusa Southeast East, Borneo South, Gorontalo, 18 province level complaint moderate health, and 9 other provinces including low levels of health complaints. It can be input to the government to pay more attention to residents in each area that has high health complaints through improving public health services so that the Indonesian population becomes healthier without exists complaint health.