Performance Evaluation of Machine Learning Approaches for COVID-19 Forecasting by Infectious Disease Modeling

The use of data analytics in virology is a rapidly growing means to provide accurate and reliable information to healthcare providers. Its mechanisms allow for deeper comprehension and characterization of pathogens (i.e., virus transmission rate and behavior). Artificial intelligence and machine learning technology have shown the potential to forecast and subdue the spread of coronaviruses. When applied to extended periods, however, the ability of these prediction models to anticipate disease spread is not promising. The development of superior algorithms is essential to improving COVID-19 forecasting accuracy. This drawback has motivated us to conduct this study with the objective of developing a COVID-19 forecasting algorithm that functions over an extended period of time. This paper highlights the mechanism of the coronavirus forecast: the Deep Learning (DL) approach. The combined utilization of online data sets and the DL approach was employed in the investigation of the life cycle and spread of COVID-19 in the Kingdom of Saudi Arabia and the Kingdom of Bahrain.