Artificial neural networks and statistical models for optimization studying COVID-19

Azhari A Elhag; Tahani A Aloafi; Taghreed M Jawa; Neveen Sayed-Ahmed; F S Bayones; J Bouslimi

doi:10.1016/j.rinp.2021.104274

Artificial neural networks and statistical models for optimization studying COVID-19

Results Phys. 2021 Jun:25:104274. doi: 10.1016/j.rinp.2021.104274. Epub 2021 May 12.

Authors

Azhari A Elhag¹, Tahani A Aloafi¹, Taghreed M Jawa¹, Neveen Sayed-Ahmed¹, F S Bayones¹, J Bouslimi²

Affiliations

¹ Department of Mathematics and Statistics, College of Science, P.O. Box 11099, Taif University, Taif 21944, Saudi Arabia.
² Department of Physics, College of Science, P.O. Box 11099, Taif University, Taif 21944, Saudi Arabia.

Abstract

The biggest challenge facing the world in 2020 was the pandemic of the coronavirus disease (COVID-19). Since the start of 2020, COVID-19 has invaded the world, causing death to people and economic damage, which is cause for sadness and anxiety. Since the world has passed from the first peak with relative success, this should be evaluated by statistical analysis in preparation for potential further waves. Artificial neural networks and logistic regression models were used in this study, and some statistical indicators were extracted to shed light on this pandemic. WHO website data for 32 European countries from 11th of January 2020 to 29th of May 2020 was utilized. The rationale for choosing the stated methodological tools is that the classification accuracy rate of artificial neural networks is 85.6% while the classification accuracy rate of logistic regression models 80.8%.

Keywords: Artificial neural networks; COVID-19; Deaths; Logistic regression; Statistical analysis.