Complex network analysis techniques for the early detection of the outbreak of pandemics transmitted through air traffic

Air transport has been identified as one of the primary means whereby COVID-19 spread throughout Europe during the early stages of the pandemic. In this paper we analyse two categories of methods – dynamic network markers ( DNMs ) and network analysis-based methods – as potential early warning signa...

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports 2023-10, Vol.13 (1), p.18174-18174, Article 18174
Main Authors: Fragua, Ángel, Jiménez-Martín, Antonio, Mateos, Alfonso
Format: Article
Language:English
Subjects:
639/166/984
639/705/1041
692/699/255
Air transportation
Aircraft
COVID-19
Disease transmission
Humanities and Social Sciences
multidisciplinary
Outbreaks
Pandemics
Science
Science (multidisciplinary)
Time series
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Air transport has been identified as one of the primary means whereby COVID-19 spread throughout Europe during the early stages of the pandemic. In this paper we analyse two categories of methods – dynamic network markers ( DNMs ) and network analysis-based methods – as potential early warning signals for detecting and anticipating COVID-19 outbreaks in Europe on the basis of accuracy regarding the daily confirmed cases. The analysis was carried out from 15 February 2020, around two weeks before the first COVID-19 cases appeared in Europe, and 1 May 2020, approximately two weeks after all the air traffic in Europe had been shut down. Daily European COVID-19 information sourced from the World Health Organization was used, whereas air traffic data from Flightradar24 has been incorporated into the analyses by means of four alternative adjacency matrices. Some DNMs have been discarded since they output multiple time series, which makes it very difficult to interpret their results. The only DNM outputting a single time series does not emulate the COVID-19 trend: it does not detect all the main peaks, which means that peak heights do not match up with the increase in the number of infected people. However, many combinations of network analysis based methods and adjacency matrices output good results (with high accuracy and 20-day advance forecasts), with only minor differences from one to another. The number of edges and the network density methods are slightly better when dynamic flight frequency information is used.
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-023-45482-9