Prediksi Analisis Penderita Covid19 di Indonesia dengan Metode Linier Regresi dan Unsupervised Learning

Abstract

The Covid-19 disease has now been declared a pandemic disease because the level of spread and the risk posed is very dangerous. Various steps such as awareness programs, social distancing, and contact tracing have been taken to control the COVID-19 outbreak. In the absence of a vaccine, prediction of confirmed cases, deaths, and recoveries is needed to increase the capacity of the health care system and control transmission. In this study, cumulative and daily cases were confirmed, died, and recovered in Indonesia. The analysis does not consider any changes in government control measures. Information from this study can provide relevant information to government and health officials and the public. How is the cure rate to the confirmed, the death rate to the number of sufferers? This study uses regression and clustering models with K-means, using unsupervised learning and supervised learning to build the distribution model. The results of this study using the regression method with R2 = 0.99 while for clustering with K = 10-15 intervals seen from the results of the elbow method.