Recurrent Neural Network and Auto-Regressive Recurrent Neural Network for trend prediction of COVID-19 in India

Abstract

Listen

Translate article

On 31st December 2019 in Wuhan China, the first case of Covid-19 was reported in Wuhan, Hubei province in China. Soon world health organization has declared contagious coronavirus disease (COVID-19) as a global pandemic in the month of March 2020. Since then, researchers have focused on using machine learning and deep learning techniques to predict future cases of Covid-19. Despite all the research we still face the problem of not having a good and accurate prediction, and this is due to the complex and non-linear data of Covid-19. In this study, we will implement RNN and Auto Regressive RNN. At first, we implement LSTM and GRU in an independent way, then we will implement deepAR with LSTM and GRU cells. For the evaluation of the obtained results, we will use the MAPE and RMSE metrics.