Future Forecasting with Machine Learning Models for Covid-19

Abstract

COVID-19 has spread around the world, putting humans at risk. Due to the widespread infection and spread of this disease, the resources of some of the world's most powerful economies are being strained. The performance of a Machine Learning prototype to predict the number of patients who will be afflicted by COVID-19, a virus that was currently contemplated a problem to humanity. In this study, four quality estimate models were used to calculate the COVID-19 problem factors: Least Absolute Shrinkage and Selection Operator (LASSO), Ridge Regression (RR), Linear Regression (LR), Random Forest (RF). For each of the models, three types of predictions are made: the count of recently cured patients,count of demise, and the count of recently affected patients. To solve the problem, a proposed method based on the long short-term Integrated Average (LSTIA) predicts the count of COVID-19 cases in the next upcoming days and the impact of preventive measures such as social distance and solitary confinement unit on COVID-19 circulation.