Pengelompokan Penyebaran Covid-19 di Provinsi Jawa Barat menggunakan Metode Clustering K-Medoids

Covid-19 is a disease that infects the human respiratory system and has a high-speed transmission ability. West Java Province is one of the areas affected by the Covid-19 pandemic. The number of people confirmed with the Covid-19 virus in West Java is still increasing daily. Therefore, it is necessary to group the level of vulnerability to the spread of Covid-19, especially in West Java Province, using data from the official website of the West Java Provincial Government using 5 attributes, namely district_city_name, total_confirmation, confirmed_recovered, confirmed_death, and confirmed_active. This study aims to identify the pattern of the spread of Covid-19 to support more effective decision-making at the regional level. The research method involves a data mining process, namely business understanding, data understanding, data preparation, modeling, evaluation, and deployment based on the CRISP-DM methodology. The modeling process uses the K-Medoids algorithm with 3 clusters according to the government's color zone. The results of this study show 3 clusters, namely the green cluster is the minimum number of cases with 16 districts/cities. The yellow cluster is starting to be alert to the number of cases with 6 districts/cities. The red cluster is a very severe case with 5 districts/cities. The results of the Silhouette Coefficient test that tested n_cluster = 2, 3, 4, and 5 showed that n_cluster = 3 is the best cluster with a value of 0.77.