Ainurrohmah, Dian Tri Wiyanti
Classification of the Covid-19 risk zone is one of the government's efforts in handling the Covid-19 pandemic in Indonesia. The division of risk zones is the basis for the government in making policies related to Covid-19 in each region. The Covid-19 risk zone is divided into four zones. The reason for this consideration was to get a calculation with the most excellent performance and get a classification pattern for the Covid-19 risk zone. The algorithm used in this study is a naive bayes, k-nearest neighbor, and decision tree with the help of WEKA software. This classification method uses 10-fold cross-validation to divide training data and test data and uses a confusion matrix to find the best performance value. The values seen to evaluate performance are accuracy, precision, recall, and time. The data used are 21190 from the official government website. The results showed that the best algorithm was a decision tree with accuracy, precision, and recall above 88%. The classification pattern generated from the best algorithm is 75 patterns. © 2023 Author(s).
Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Negeri Semarang, Jalan Raya Sekaran, Central Java, Semarang, Indonesia
