Deep Learning Model for Predicting Spreading Rates of Pandemics, “COVID-19 as Case Study”

Abstract

The outbreak of Coronavirus (COVID-19), especially SARS-CoV-2, has led to a catastrophic scenario in the course of the world. The cumulative prevalence of COVID-19 was increasing rapidly day by day. Machine learning (ML) and deep learning (DL) can be deployed to facilitate tracking disease, anticipating the increase in epidemic, and hence planning for coverage techniques to control its spread. This work is based on the application of an advanced mathematical model to examine and predict the increase in a pandemic. On the bases of time-series data, an advanced DL model has been implemented to predict the risk of COVID-19 spreading in Iraq. A hybrid approach is presented where two deep learning algorithms; LSTM and GRU are brought up together to achieve good prediction with rewarding levels of (MAE = 0.109), (MAPE = 0.191) and (RMSE = 0.134).