Prediksi Jumlah Pasien Covid-19 Dengan Menggunakan Klasifikasi Algoritma Machine Learning

Abstract

Corona virus or servere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a disease that results in the occurrence of mild to moderate respiratory tract infections. Positive cases of Covid-19 in Indonesia were first detected on March 2, 2020 and continue until 2022. The additional number of deaths caused by COVID-19 has also increased. Therefore, the author is interested in making a predictive model of the cumulative number of COVID-19 patients who died in Indonesia. Therefore, in this study is how to predict the number of patients who die from COVID-19 in Indonesia by creating an appropriate accuracy model to help estimate the number of deaths associated with COVID-19 in Indonesia and assist the government in dealing with cases of new variants of COVID-19. In this study, the authors used the Decision Tree model using entropy criteria as well as Information Gain and Random Forest which resulted in accuracy rates of 91.83% (Decission Tree) and 73.80% (Random Forest). The results, explain that the model used is good. The more the R-squared error value is close to 1, the better the model used will be