Labeling Indonesia COVID-19 Data Using K-Means Clustering with Optimization

Abstract

COVID-19 or Corona Virus Diseases is a virus that spreads throughout the world and causes a pandemic that affects social life, education and tourism, especially in Indonesia. The government has implemented various policies to reduce the rate of cases in Indonesia. In determining policies and regulations, the role of data is very important, especially in Indonesia, but the existence of data is still small and has not been labeled. In this study, the method used to label COVID-19 data in Indonesia is using K-Means Clustering. K-means is a data processing method that produces a group that is divided into 16,936 data. Determination of the number of groups in this study using the Elbow method and optimized by the Davies Bouldin Index method. The result of this study is the number of clusters used as labeling of COVID-19 data in Indonesia. The number of clusters was obtained using the Elbow method and optimized with the Davies Bouldin Index so as to produce a total of 4 clusters and the results of the labeling obtained the number of members in each cluster which amounted to 15315 in cluster 0, 1191 in cluster 1, 222 in cluster 2 and 208 in cluster 3.