Hybrid Machine Learning Model Coupled with School Closure For Forecasting COVID-19 Cases in the Most Affected Countries

Coronavirus disease (Covid-19) caused millions of confirmed cases and thousands deaths worldwide since first appeared in China. Forecasting methods are essential to take precautions early control the spread this rapidly expanding pandemic. Therefore, research, a new customized hybrid model consisting Back Propagation-Based Artificial Neural Network (BP-ANN), Correlated Additive Model (CAM) Auto-Regressive Integrated Moving Average (ARIMA) models were developed forecast Covid-19 prevalence Brazil, US, Russia India. dataset is obtained from World Health Organization website 22 January, 2020 6 2021. Various parameters tested select best ARIMA for these countries based on lowest MAPE values (5.21, 11.42, 1.45, 2.72) India, respectively. On other hand, proposed BP-ANN itself provided less satisfactory values. Finally, was achieved obtain results (4.69, 6.4, 0.63, 2.25) forecasting Those emphasize validity our model. Besides, prediction can assist terms taking important world.