Analysis and Prediction of COVID-19 Confirmed Cases Using Deep Learning Models: A Comparative Study

Abstract

COVID-19 or Novel coronavirus is an infectious disease that was first noticed in December, 2019 and it eventually emerged as a pandemic as it is highly contagious in nature. It affected the economic and social structure worldwide and caused a huge loss of human life. Due to the scarcity of medical infrastructure, it has become nearly impossible to cure every case of COVID-19 and hence the loss of lives is exceedingly increasing. So, if the cases can be forecasted beforehand, proper precautions can be taken on time and thousands of human lives can be saved. In this paper, predictions of the number of coronavirus confirmed cases for the five topmost affected countries across the world have been made. Along with it, a comparative study of ANN (Artificial Neural Network) and RNN (Recurrent Neural Network) based LSTM (Long Short Term Memory) Model has been carried out. The countries taken into consideration for this paper are USA, India, Brazil, Russia, and France. The models have been used to train the dataset and validate the prediction results against the original data based on the predefined metric of MSE or Mean Squared Error. The prediction results have been visualized graphically and it was inferred that the LSTM model outperformed the ANN model.