Aplikasi Data Mining untuk Clustering Daerah Penyebaran Penyakit Diare di DKI Jakarta Menggunakan Algoritma K-MEANS

Abstract

Diarrhea is a disease that makes sufferers defecate more than 4 times, diarrhea usually strikes all people, including toddlers, teens and adults. The increasing population density in DKI Jakarta Province in 2017 around 10,177,924 inhabitants spread over 6 cities consisting of 44 districts and 267 villages. To be able to see the area of diarrhea distribution, it is necessary to make a grouping based on the attributes used consisting of JSP, JSSP, SHARING and OD in order to obtain the center of the diarrhea spread point. K-Means algorithm is very suitable in clustering the spread of diarrhea Kelurahan. To determine the centroid starting point based on diarrhea sufferers as a parameter in determining C1, C2 and C3 which get C1 is Duren Tiga Village, C2 is Kebon Baru Village and C3 is Bendungan Hilir Village. The results of grouping based on the attributes used get C1 totaling 11 Kelurahans, C2 totaling 34 Kelurahans and C3 totaling 43 Kelurahans from 88 Kelurahan data used in which C1 is categorized as a diarrhea prone point, C2 is categorized as a possible spread of diarrhea and C3 is categorized as a Kelurahan that is safe from spread of diarrhea.