A Review of Various Mathematical and Deep Learning based Forecasting Methods for COVID-19 Pandemic

ABSTRACT

With over a hundred million cases worldwide and thousands coming daily, the outbreak of COVID-19 has seriously affected many countries' healthcare and economic situations. A precise and efficient model for predicting new COVID-19 cases and the pandemic's future dynamics can be highly beneficial in such distressing conditions. These predictions might help the hospitals and the concerned authorities to devise necessary and preliminary arrangements for the patients in advance. This will be able to positively prevent the second or third wave of the pandemic spread. In the following study, we have composed a brief analysis of the appropriate and recent tools used for forecasting COVID-19. In this study, we have categorized these forecasting techniques into two broad classes, viz. Mathematical modeling based and Deep Learning-based. These predictions prepare us against any future threat and consequence that may occur in the future. © 2021 IEEE.