Forecasting of COVID-19 Using Supervised Machine Learning Models

Abstract

Machine learning (ML) models have proved significant in forecasting to improve decision-making. Various application domains, including the identification of adverse factors for a hazard, have long used machine learning models. To forecast the problems, several prediction approaches have been used. For a long time, machine learning algorithms have been used in a variety of applications, including detecting negative risk factors. This research demonstrates the ability of machine learning models to predict the number of patients who would be infected by COVID-19, a virus that may pose a danger to humanity. The three standard forecasting models used in this study were Linear Regression, Support Vector Machine (SVM), and Exponential Smoothing. In the next 20 days, each of these models has different types of forecasts, such as cases that are confirmed newly, new deaths, and new recoveries predictions. The results of the study suggest that these models are better used in the most recent COVID-19 analysis. ES performs better all other ones.