Analisis Algoritma K-Means dalam Pengelompokkan Persebaran Covid-19 di Indonesia

Abstract

Covid-19 or Coronavirus is a virus that is found in humans and animals. This virus can infect humans to cause various diseases such as flu, to serious diseases such as Middle East Respiratory Syndrome (MERS) and Severe Acute Respiratory Syndrome (SARS). In Indonesia, the spread of Covid-19 cases continues to increase and is evenly distributed in all provinces in Indonesia because of the fairly rapid spread due to the vast area in Indonesia, making it possible for grouping based on regions in Indonesia to be needed which will result in the center points of the spread of this Covid-19 case. This study aims to group Covid-19 data into a cluster using the K-Means Clustering Data Mining Algorithm. The Covid-19 data used in this study is Covid-19 data on July 6, 2021 which was taken from the official website of Kawal Covid-19 (KawalCovid-19.id). The attributes used are positive cases, recovered, and died. The clusters formed from the results of research using K-Means Clustering are 3 clusters with the first cluster consisting of 2 provinces, the second cluster 3 provinces, and for the third cluster 29 provinces. The cluster with the largest Covid-19 spread rate is cluster one. From this study, the accuracy was 91.176% and evaluated using the Davies-Bouldin Index yielded a fairly good cluster result with a value of 0.493371469.