PERBANDINGAN ANALISIS CLUSTERING K-MEANS DAN K-MEDOIDS PADA DATA PENYAKIT DI INDONESIA TAHUN 2019

Abstract

The spread of disease in Indonesia is a serious problem and must be addressed. Provinces in Indonesia have different characteristics of the spread of disease in each region. Characteristics of an area are grouped based on indicators of disease spread, so that the government can accurately and quickly take disease prevention policies in an area by grouping. This study discusses the grouping of provinces in Indonesia based on disease cases in 2019 using a comparison of the K-Means and K-Medoids clustering methods which include non-hierarchical data grouping methods. The results of this study indicate that using K-Means obtained 3 provinces in cluster 1, 29 provinces in cluster 2 and 2 provinces in cluster 3, while using K-Medoids obtained 29 provinces in cluster 1, 4 provinces in cluster 2 and 1 province. in cluster 3. From the results of grouping the two methods, a comparison of the best method using cluster validation is obtained, namely the K-Means method because it has the smallest variance value, which is 1.010.