Prediction of the COVID-19 pandemic with Machine Learning Models

ABSTRACT

The latest destructive outbreak, Corona virus (2019), is rapidly sweeping the globe. Not only are economies deteriorating, but countries' entire strengths and confidence are as well. Machine learning forecasting strategies have demonstrated their importance to anticipate in outcomes of the perioperative period to improve the future decision-making actions. The machine learning algorithms have long been used in several applications which require the detection of adverse factors for a threat. Forecasting techniques are essential for producing accurate results. This study shows the ability to predict the number of cases affected by COVID-19 as potential risk to mankind. In this analysis, four-prediction algorithms have been used which are linear regression (LR), Exponential Smoothing (ES), least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM). Each of these models has three different kinds of predictions, such as the newly infected patients, death cases and the recovery cases in the next ten days. These approaches are better used to forecast the covid-19 pandemic, as shown by the findings of analysis. The ES, that is effective in forecasting new corona cases, death cases and recovery cases.