Applying the ARIMA Deep Learning Algorithm to Predict the Coronavirus in the Kingdom of Saudi Arabia using Time Series Data

The COVID-19 pandemic has posed a significant threat to humanity, with irreversible societal consequences. Research is being conducted to anticipate the development or return of the pandemic at any later time and discover more effective vaccinations, and consequently reduce the death toll. Maybe in the future context, accurate COVID-19 prediction using deep learning is gaining increasing attention as deep learning approaches are more successful in dealing with non-linear situations. Time series prediction of COVID-19, in terms of the estimated number of confirmed, death, and recovered cases, is performed in our study utilizing the ARIMA model. Short-term infected cases are all predicted in the proposed methodology. We used daily data from April 1, 2020, to May 31, 2021, to train and evaluate the models for our study. The models used in this study are data-driven, and we use two MAPE, and R2 metrics to assess our models? predictive performances. We aim to evaluate and contrast the abilities of ARIMA the model in interpreting complex time series trends, and ultimately forecasting new cases for the future period of 14 days. Our methods and predicted consequences will aid in the prevention of COVID-19 pandemic infections.