Covid-19 Confirmed-Cases Prediction in SAARC Countries through Machine Learning

In December 2019, a new variant of the SARS virus named Severe Acute Respiratory Syndrome Coronavirus 2(SARS-CoV-2), began its outspread in Wuhan, China, and has since expanded throughout the whole planet. In this novel research, we predicted the number of confirmed cases of SARS-CoV-2 in the South Asian Association for Regional Cooperation (SAARC) Countries through the use of the numerous machine learning (ML) techniques and time series model. Furthermore, we made a comparative study on which technique performed better. The hugely popular Support Vector Machine(SVM) and Bayesian Ridge regression was taken into consideration for the predictions made. The time series analysis model, i.e. Seasonal Autoregressive Integrated Moving Average(SARIMA) model was further used to get even better predictions on the forecasts for the confirmed cases of SARS-CoV-2. Along with this, comparisons were conducted on the confirmed cases, followed by the deaths resulted from these cases, as well as on the number of recoveries made, the number of active cases, and mortality rate across these countries, from which nations were limited down to a handful that should take extreme steps to stop the virus from spreading.