Data Mining using clustering method to predict the spread of Covid 19 based on screening and tracing results

Abstract

Coronavirus is a virus that causes disease in humans and animals. The virus was discovered in Wuhan, China in December 2019. Initially, it was suspected to be pneumonia, with general symptoms similar to the flu. However, unlike influenza, coronaviruses can progress rapidly, leading to more severe infections and organ failure. The number of COVID-19 sufferers in Indonesia is increasing every month. Anticipation and reducing the number of people infected with the coronavirus in Indonesia have been carried out in all regions. Including providing policies that limit activities outside the home. Indonesia has a very wide area, so it is necessary to classify the spread of Covid-19 based on regions or regions in Indonesia. This grouping provides a central point for the spread of Covid-19 pandemic cases in Indonesia. In testing data using data mining, data mining allows users to find knowledge in databases that were previously unknown to the user. By using the Clustering technique and the K-Means algorithm to predict the spread of COVID-19 based on the results of screening and tracing. The Clustering method produces 3 clusters, Cluster 0 with a medium category with a total of 6 regions, Cluster 1 with a low category with a total of 3 regions, and Cluster 2 with a high cluster with a total of 7 regions, with a DBI value of -0.784.