COVID-19 Cases Estimation in the UK Using Improved SEIR Models

Jiang, Shaogang, Al-Ataby, Ali and Al-Naima, Fawzi (2021) COVID-19 Cases Estimation in the UK Using Improved SEIR Models. In: 2021 14th International Conference on Developments in eSystems Engineering (DeSE), 2021-12-7 - 2021-12-10, Sharjah, United Arab Emirates.

Text Paper_IEEE_format_revised - 17-10-2021.docx - Author Accepted Manuscript Download (395kB)

Abstract

The paper suggests a machine learning algorithm with two modified SEIR models customized for the 2019-nCoV virus and vaccine uses to simulate the spread of COVID-19 in the UK (from Jan 2020 to March 2021) and make predictions of future cases. The algorithm uses COVID daily cumulative case data and second dose vaccine use data provided by the Public Health England as the training set and is capable of making relatively accurate short-term predictions of future COVID cases in the UK (before the delta and later variants of the virus starts spreading within the country). The obtained overall accuracy is above 80% for daily incremental case numbers in terms of the overall fit of the model to real-life data, and with an accuracy of more than 80% for estimation of daily incremental case numbers for 14 days period future prediction. The goal of this paper is to propose improved SEIR models capable of a more accurate simulation for COVID-19 modelling and estimation with various machine learning algorithms.

Item Type:	Conference or Workshop Item (Unspecified)
Uncontrolled Keywords:	46 Information and Computing Sciences, 4611 Machine Learning, Bioengineering, Emerging Infectious Diseases, Vaccine Related, Infectious Diseases, Coronaviruses, Machine Learning and Artificial Intelligence, Networking and Information Technology R&D (NITRD), Immunization, 3 Good Health and Well Being
Divisions:	Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science
Depositing User:	Symplectic Admin
Date Deposited:	01 Dec 2021 12:02
Last Modified:	08 Dec 2024 01:13
DOI:	10.1109/dese54285.2021.9719390
Related URLs:	Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3144334