Topics and Sentiments of Public Concerns Regarding COVID-19 Vaccines: Social Media Trend Analysis

Abstract

Background As a number of vaccines for COVID-19 are given emergency use authorization by local health agencies and are being administered in multiple countries, it is crucial to gain public trust in these vaccines to ensure herd immunity through vaccination. One way to gauge public sentiment regarding vaccines for the goal of increasing vaccination rates is by analyzing social media such as Twitter. Objective The goal of this research was to understand public sentiment toward COVID-19 vaccines by analyzing discussions about the vaccines on social media for a period of 60 days when the vaccines were started in the United States. Using the combination of topic detection and sentiment analysis, we identified different types of concerns regarding vaccines that were expressed by different groups of the public on social media. Methods To better understand public sentiment, we collected tweets for exactly 60 days starting from December 16, 2020 that contained hashtags or keywords related to COVID-19 vaccines. We detected and analyzed different topics of discussion of these tweets as well as their emotional content. Vaccine topics were identified by nonnegative matrix factorization, and emotional content was identified using the Valence Aware Dictionary and sEntiment Reasoner sentiment analysis library as well as by using sentence bidirectional encoder representations from transformer embeddings and comparing the embedding to different emotions using cosine similarity. Results After removing all duplicates and retweets, 7,948,886 tweets were collected during the 60-day time period. Topic modeling resulted in 50 topics; of those, we selected 12 topics with the highest volume of tweets for analysis. Administration and access to vaccines were some of the major concerns of the public. Additionally, we classified the tweets in each topic into 1 of the 5 emotions and found fear to be the leading emotion in the tweets, followed by joy. Conclusions This research focused not only on negative emotions that may have led to vaccine hesitancy but also on positive emotions toward the vaccine. By identifying both positive and negative emotions, we were able to identify the public's response to the vaccines overall and to news events related to the vaccines. These results are useful for developing plans for disseminating authoritative health information and for better communication to build understanding and trust.

Social media and attitudes towards a COVID-19 vaccination: A systematic review of the literature

“…Namely, the peaks of the topics were affected by the events reported by the media and posted on social media. 60 Further, vaccine-supportive tweets showed temporal variations over time, while those related to barriers remained mainly constant through time. 61 After conducting the updated search,we were able to locate new studies that focused social data mining on specific regions, thereby providing more insights into themes that were specific to a certain location.…”

Sentiment analysis of Indian Tweets about Covid-19 vaccines

“…The study shows that Twitter has evolved into a vital source of information and can be utilised in real-time situations to manage various crises by properly analysing the data. By identifying the sentiments, public reactions about the Covid-19 vaccine can be determined, which will aid in developing programmes for providing authoritative health information and improving communication to foster understanding and confidence [ 83 ]. However, it is disconcerting that the negative sentiment towards vaccines rests with some people in India.…”

Fear of Infection and Sufficient Vaccine Reservation Information Might Drive Rapid Coronavirus Disease 2019 Vaccination in Japan: Evidence from Twitter Analysis

“…In previous works of large-scale Twitter analyses, after preprocessing, there are mainly four types of NLP methods: n-gram token analysis [12,15,20,21], sentiment analysis [12,14,15,[20][21][22][23][24][25], topic modelling [12,14,15,20,[22][23][24][25], and geographical analysis [22,24]. The geographical analysis is less important in our work because the range of our research is a whole country instead of subareas.…”

“…After n-gram analysis, sentiment analyses were often used to explore the real-time public attitudes in social media analysis related to COVID-19 vaccination, which may reflect the acceptance of COVID-19 vaccines and related policies [12,14,20,22,24]. Trend of negative sentiments may provide potential evidence for vaccine hesitancy [23]. In this work, sentiment analysis was applied to all vaccine-related tweets.…”

“…Based on the sentiment analysis results, we summarized the topics to look deeper into the dominant emotion in the tweets. Latent Dirichlet allocation (LDA) is often used in tweets topic modeling studies [14,15,20,22,23]. In this study, LDA regards tweets as being generated from different topics, and each topic generates tweets with a Dirichlet distribution.…”