Analysis of COVID-19 Vaccine Brand Sentiment on Twitter
Date Issued
Creators
Journal Title
Journal ISSN
Volume Title
Publisher
Northern Kentucky University
Abstract
"During March and April, 2021, we collected tweets that contain both a COVID keyword and a vaccine keyword, using the Twitter free streaming API. In this work we focus on analyzing the tweets collected during the week, April 1 – April 7. Each tweet was placed in one or more subsets of tweets based on particular vaccine brand keyword that it contains. Using this approach, we created eight tweet subsets (non-disjoint) for the following COVID vaccines:‘Johnson and Johnson’, ‘AstraZeneca’, ‘Pfizer’, ‘Moderna’, ‘Novavax, ‘Sputnik’, ‘Sinovac’, and ‘Covaxin’. For each subset, we computed the number of tweets per unit of time and the average sentiment score of the tweets during that time window, using a publicly available lexicon-based technique (TextBlob). We visualize both the count and
average sentiment values for all 8 vaccine types."
Description
2021 Celebration of Student Research and Creativity presentation
Type
Presentation
Format
Keywords
COVID-19 (Disease), Social media in medicine, Computer science