No Detail Found!

Public opinion, clusters, and statistics of Covid-19 vaccination: a twitter analysis

Author(s)
GHADA AMOUDI Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia
ZELAL SHEARAH Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia
Abstract
The coronavirus disease (COVID-19) pandemic has drastically affected the entire world. Vaccinations have been developed to contain the spread of the virus and help humanity recover from the pandemic. However, people are often reluctant to adopt new medical interventions, and COVID-19 vaccines are no exception. Social media platforms like Twitter are flooded with discussions, opinions, and rumors about vaccines. Numerous people access news regarding COVID-19 vaccines on Twitter-rather than through official media channels-which offers a wealth of comments about news and medical breakthroughs. However, as people are panicked due to this alarming situation, they tend to share any information they receive without checking its credibility, creating more panic. Analyzing interaction between social media users provides insight into the spreading pattern of news, people's opinions towards a particular issue, who the influencers are, and how they influence others, among other findings. This study investigated opinions about COVID-19 vaccines by applying sentiment analysis and detecting communities on Twitter. We constructed an interaction network of discussions related to COVID-19 vaccines and applied social network analysis methods to find communities and nodes with high centrality measures. Next, we analyzed how these nodes affect the overall community opinion. Two main communities were detected, with the larger community displaying a higher positive sentiment ratio than the smaller one. Furthermore, the polarity of the high centrality nodes in each community was close to the average polarity of the community as a whole. These findings highlight the potency of the node's position in terms of centrality measures. In conclusion, analyzing discussion networks should not be overlooked when public health is concerned, as influencers are not necessarily those with high numbers of followers but rather those with high centrality measures within the interaction network regarding the topic being discussed.
Publication Details
Page(s) 6036-6048
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 20, Year: 2022

Request the full text from PASTIC

Keywords
Betweenness, Sentiment analysis, Girvan–Newman, Centrality, Community Detection
0

Citations

0

Downloads

114

Views