Comparison of Machine Learning Approaches for Anticipating of COVID-19 Active, Recovered and Death Cases in India

ABSTRACT

Outbreaks of the COVID-19 that emanated in Wuhan city of China have been causing worldwide health concerns since December 2019 resulting in a global pandemic declared by World Health Organization (WHO) on March 11, 2020. It has highly affected social, financial matters and health too. In the study, COVID-19 affected people’s statistics are taken into account for predicting the upcoming day’s movement in a total count of infected cases in India. Regression models especially multiple linear regression, support vector regression are implemented on the dataset for observing the curve of the infected cases and forecast the total active, total deaths and total recovered cases for next coming days. The usefulness of regression techniques is studied. These techniques analyze and predict the rise and spread of COVID-19. We investigate how well mathematical modeling can forecast the rise using datasets from https//covid19india.org. Here, a comparison of multiple regression and support vector regression is done. It can be concluded that these models acquired remarkable accuracy in forecasting COVID-19. We also want to compare the distribution of COVID-19 in different nations and try to predict potential instances as soon as possible. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.