Prediction of covid-19 daily infected cases (worldwide & united states) using Regression Models and Neural Network

Abstract

Covid-19 is an infectious disease that has threatened the world by spreading at an alarming rate. Since the start of this pandemic, various researchers are working on the prediction of infection rates, death rates, daily cases at the country level as well as worldwide. Researchers are using different machine learning techniques to predict these values. However, the choice of appropriate features is a very important task for accurate predictions. According to literature, different variables like covid-19, demographic and climate play a significant role in these predictions. This project is focused on the prediction of daily cases using linear regression, polynomial regression and a 2 layer neural network with and without regularization. The study deals with the prediction of cases in United states at state level as well worldwide cases.