Open AccessSocial network analysis and community detection on spread of COVID-19
Author(s) -
Ashani Wickramasinghe,
Saman Muthukumarana
Publication year - 2021
Publication title -
model assisted statistics and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.178
H-Index - 11
eISSN - 1875-9068
pISSN - 1574-1699
DOI - 10.3233/mas-210513
Subject(s) - centrality , exponential random graph models , social network analysis , covid-19 , computer science , coronavirus , social network (sociolinguistics) , econometrics , network analysis , index (typography) , community structure , graph , data science , random graph , data mining , statistics , mathematics , theoretical computer science , social media , world wide web , medicine , physics , disease , pathology , quantum mechanics , infectious disease (medical specialty)
This paper explains the epidemic spread using social network analysis, based on data from the first three months of the 2020 COVID-19 outbreak across the world and in Canada. A network is defined and visualization is used to understand the spread of coronavirus among countries and the impact of other countries on the spread of coronavirus in Canada. The degree centrality is used to identify the main influencing countries. Exponential Random Graph Models (ERGM) are used to identify the processes that influence link creation between countries. The community detection is done using Infomap, Label propagation, Spinglass, and Louvain algorithms. Finally, we assess the community detection performance of the algorithms using adjusted rand index and normalized mutual information score.