COVID-19 future forecasting using supervised machine learning models

Abstract

The spread of COVID-19 in the entire world has put the humankind in danger. The assets of probably the biggest economies are worried because of the enormous infectivity and contagiousness of this illness. The ability of ML models to conjecture the quantity of forthcoming patients influenced by COVID-19 which is by and by considered as a likely danger to humanity. Specifically, four standard estimating models linear regression (LR), least total shrinkage and determination administrator (LASSO) Support vector Machine (SVM) have been utilized in this examination to figure the undermining components of COVID-19. Three sorts of expectations are made by every one of the models, for example, the quantity of recently tainted cases, the quantity of passing, and the quantity of recuperations But in the can't foresee the precise outcome for the patients. To defeat the issue, Proposed strategy utilizing the exponential smoothing (ES) anticipate the quantity of COVID-19 cases in next 30 days ahead and impact of preventive estimates like social seclusion and lockdown on the spread of COVID-19. © 2021 Karadeniz Technical University. All rights reserved.