• Users Online: 130
  • Print this page
  • Email this page


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2022  |  Volume : 3  |  Issue : 1  |  Page : 11-14

Studying public perception about covaxin vaccination: A sentiment analysis of tweets


1 Research Scholar SRM School of Public Health, SRM institute of Science and Technology, Kanchipuram, Tamil Nadu, India
2 SRM School of Public Health, SRM Institute of Science and Technology, Kanchipuram, Tamil Nadu, India
3 Division of Gastrointestinal Sciences, The Wellcome Trust Research Laboratory, Christian Medical College, Vellore, Tamil Nadu, India
4 Care India Foundation, Chennai, Tamil Nadu, India

Date of Submission29-May-2021
Date of Decision08-Aug-2021
Date of Acceptance22-Sep-2021
Date of Web Publication25-Mar-2022

Correspondence Address:
Dr. Saravanan Chinnaiyan
Research Scholar SRM School of Public Health, SRM institute of Science and Technology, 3rd Floor, Kanchipuram, Tamil Nadu
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/jphpc.jphpc_13_21

Rights and Permissions
  Abstract 


Background: After a year of the COVID-19 pandemic began, there are approximately 200 vaccine candidates in development. Ten of them have been approved by several countries or used in limited emergency situations. India is gearing up to launch its vaccination campaign on January, 16. Vaccination attitudes can significantly influence individual vaccination decisions. Measuring such feelings and their distribution in a population is a time-consuming and challenging task. Our objective of the study is to analyze the current sentiment of Covaxin vaccination on social media on Twitter. We have retrieved 4139 tweets posted from March 1, 2021, to March 31, 2021, by the Twitter program interface. Results: Our study results show that 38.8% of the respondents have neutral emotions toward the Covaxin vaccination, 35.4% believe positive, and 25.8% are negative have a negative perception. There was no significant association between tweets sentiments and users (P = 0.1976) at 0.05 level of significance. Conclusions: In the internet age, unsubstantiated vaccine safety concerns, the impact of fast rumors, and misinformation can spread quickly. It's up to policymakers to interpret the general questions as a plausible public reaction. Attempts should be made to combat vaccine misinformation through various platforms, such as newspapers, television advertisement, and social media campaign.

Keywords: COVAXIN, COVID-19, sentimental analysis, twitter, vaccination


How to cite this article:
Chinnaiyan S, Govindaraj YK, Dharmaraj A, Babu B. Studying public perception about covaxin vaccination: A sentiment analysis of tweets. J Public Health Prim Care 2022;3:11-4

How to cite this URL:
Chinnaiyan S, Govindaraj YK, Dharmaraj A, Babu B. Studying public perception about covaxin vaccination: A sentiment analysis of tweets. J Public Health Prim Care [serial online] 2022 [cited 2022 May 27];3:11-4. Available from: http://www.jphpc.com/text.asp?2022/3/1/11/340806




  Introduction Top


India launched its national SARS-CoV-2 vaccination program, responsible for the COVID-19 pandemic, on January 16, 2021. The priority will be health care and front-line workers older than 50 years or who suffer from certain medical conditions. According to health officials, India has administered 3,950,156 vaccine doses across the country as of February 1, 2021.[1]

As of late February, the Serum Institute of India began animal testing of vaccine candidates, followed by Zydus Cadila in March. In May, ICMR partnered with Bharat Biotech to develop a complete COVID vaccine in India. More than 30 COVID-19 vaccine candidates were in development in India until May, many of whom were already in preclinical tests. Per reports that emerged in July, the ICMR was preparing to launch the BBV152 COVID or Covaxin vaccine.

Covaxin is made from an inactivated SARS-COV-2 vaccine (BBV152), an aluminum hydroxide gel adjuvant (Algel), or a novel Algel adsorbed agonist TLR7/8. The inactivated vaccine formulation containing TLR7/8 adjuvant-induced agonist Th1 partial antibody response increased IgG2a/IgG1 ratio and increased SARS-CoV-2 specific interferon-Δ+ CD4 T lymphocyte response.[2]

India's Health Minister announced in September 2020 that the first vaccine for use would be available from the first quarter of 2021. Nearly thirty million health workers directly dealing with COVID patients are expected to be the first to receive the vaccine, particularly doctors and other medical staff.[3]

The national vaccine drive was begun on January 16, 2021 through 3006 vaccine centers. Every vaccination center is provided with Covishield or Covaxin, but not both. Few states have decided to use only Covishield as a default option and keep their Covaxin doses as a buffer stock.[4] Since Covaxin has not completed phase 3 trials, those receiving Covaxin will need to sign a signed consent form. Immunization occurs through sociocultural, political influences, and communications responses are locally required to achieve the objectives coverage and maintain them.

On the 1st day (16 January), 165,714 people were vaccinated. In the first 3 days, 631,417 people have been vaccinated. Of these, 0.18% reported side effects, and nine (0002%) admitted to hospitals for observation and treatment.

Sentiment analysis is a process of natural language processing (NLP) to extract attitudes, opinions, views, and emotions from text, speech, tweets, and database sources (NLP). Sentiment analysis is the process of categorizing textual thoughts into categories such as “positive,” “negative,” or “neutral.”[5] Subjectivity analysis, opinion mining, and appraisal extraction are other terms for it.

The Internet medium has changed the way people express their thoughts and opinions today. It is now primarily accomplished through blog posts, online forums, product review websites, social media, and other similar mediums.[6] Millions of people use social media such as Facebook, Twitter, Google Plus, and others to express their emotions, share opinions, and share perspectives about their daily lives.[7] We get interactive media through online communities, where consumers inform and influence others through forums. In the form of status updates, tweets, blog posts, comments, and reviews, social media generates a large volume of sentiment-rich data.[8]

Every day, millions of people use Twitter, one of the most popular social media platforms globally, to share their thoughts on various topics.[9] Around 10,000 tweets are sent out every second, for a total of 700 million per day. Furthermore, Twitter restricts its users to a maximum of 140 characters per post. This constraint forces users to share their thoughts in a very concise manner. Twitter is a valuable and rich data source for analyzing public opinion because of its massive number of concise tweets.[10]

In the last few years, the growth of social media has created new opportunities to measure health behavior. Recently, the NIPAH outbreak in the Kerala virus has demonstrated the power to use web data to monitor events in real-time, but the evaluation of health behavior has remained inadequate.[11] Here, we used short texts from an online social service that was publicly available (Twitter).


  Materials and Methods Top


The data have been collected through the twitter Application programming interface (API). We searched for all public tweets (excluding retweets) posted from March 1, 2021 to March 31, 2021 that mentioned #COVAXIN #BHARATBIOTECH, with language as English. The search returned to 2500. For each tweet, we stored the text and some related metadata (tweet ID, user ID, user geolocation, number of favorites or likes, language).

The polarity analysis was performed based on the lexicon-based model called vader_lexicon, which annotated a simple random sample of original tweets for sentiment polarity such as positive, negative, or neutral.

Tweet ID and user ID have been anonymized with data masking, and the usernames mentioned in the tweets have been replaced with the user mention code. We used the user geolocation and the user-reported profile location to determine each tweet's place at the country level.

Text preprocessing and data cleaning were performed in all the tweets by the following steps:



Statistical analysis

The descriptive analysis (emotions of positive, negative, and neutral) and cumulative distribution function of the tweets has been performed. Word cloud is created to determine its relative importance quickly and with the most prominent terms. To determine relationship, a correlation analysis was carried out.


  Results Top


Sentiment distribution

We have retrieved 4139 from Public tweets. According to the funnel chart of sentiment distribution, 38.8% of the respondents have neutral emotions toward Covaxin vaccination, 35.4% felt positive to Covaxin, and 25.8% has a negative perception [Figure 1].
Figure 1: Funnel chart of sentiment distribution of the tweets

Click here to view


Distribution of sentiments based on their locations

People are sharing their tweets from different parts of the world. We have recorded the geolocation of each tweet. The top positive, negative, and neutral tweets about their location are interpreted in [Table 1].
Table 1: Distribution of sentiments based on their locations

Click here to view


Most commonly used words in tweets

Using text mining analysis, we have processed every character of the tweet. The most commonly used words are described in [Table 2].
Table 2: Most commonly used words in tweets

Click here to view


Cumulative distributive function

We have performed the cumulative distribution of the function of our tweets. The results stated that the positive and negative sentiments distributions of the sentiments follow a normal distribution. Therefore, there may be no significant differences in the strength of our data [Figure 2] and [Figure 3].
Figure 2: Distribution of sentiments across our tweets

Click here to view
Figure 3: Cumulative distribution of tweets

Click here to view


Correlation

Pearson correlation and Spearman correlation were performed to for describing the relationship. We assume P = 0.05 as a level of significance. There was no statistically significant association between tweets sentiments and users (P = 0.1976) [Figure 4].
Figure 4: Correlation analysis of sentiments and users

Click here to view


Word cloud

A word cloud is a visual representation of the frequency of words in a collection. The larger the term appears in the image generated, the more frequently it appears in the text being analyzed [Figure 5].
Figure 5: Word cloud

Click here to view



  Discussion Top


We found that the public opinion about the COVAXIN vaccine is almost equal in the distributions of emotions from March 1, to March 31, 2021. The repercussions of the COVID-19 pandemic were recognized by provaccinators, who backed vaccination research. Misinformation and adverse consequences, which are statistically rare, influenced vaccine sceptics.

This is one of only a handful few investigations to analyze online sentiment, attitudes, and word usage on the COVID-19 vaccine using the twitter application programming interface (API). We distinguished the data in various strategies including, user Geolocation, number of favorites or likes, language, etc.[12] The main target is on the details regarding the COVID-19 vaccine. This technique may also be applied to alternative domains and areas of health. The polarity analysis was performed based on the lexicon-based model called vader_lexicon.[13] The findings of this study mainly focuses on the related topics OF COVID-19 vaccines and the most commonly used words on the tweets. The percent of neutral sentiment tweets were more when compared to the tweets with the positive and the negative sentiment. The most commonly used words in tweets were found to be a phase, neutralization, seeking and serum, etc. The commonly used tweet words were found separately for the positive, negative, and neutral sentiments tweets. In this context, all the tweets were analyzed, no matter whether or not they originated from one or multiple users. Thus, it portrays the public issue on various problems associated with disease and its prevention, as a consequence imparting a lens into the extent of focus of public health.

It is thrilling to discover elements that could make contributions to the web-based posting of terrible sentiments. An empirical study of Facebook users revealed that positive information spreads quickly but does not last as long as negative information.[14] In this context, upcoming research can examine whether or not there is a choicest duration wherein facts may be supplied online to create a high-quality affect and preserve it lively in memory. It is likewise really well worth reading whether or not humans submit negative sentiments on vaccines simply as an attention-in search of gesture of providing significantly differing opinions. Particularly in health care, its miles really well worth searching at a method to inspire humans with high-quality sentiments to stay lively and contribute greater online.[15] People are more likely to favor long-term gains over short-term costs when experiencing positive feelings. Finally, upcoming research can include techniques entailing different platforms to be able to correlate sentiment tweets across age groups and across platforms.


  Conclusions Top


Vaccine hesitancy is a complex issue. Vaccine hesitancy reduces immunizing agent uptake and compromises herd immunity. The “3 Cs” such as confidence, self-complacency, and convenience models should be used for the awareness program. Political leaders and religious leaders should open up support for vaccination drives and consider that religious texts and beliefs do not oppose disease prevention. More education materials are needed, and their claims of integrity and ethics are updated, and specific data for the public and vaccination strategies should be contextualised.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Bhuyan A. India begins COVID-19 vaccination amid trial allegations. Lancet 2021;397:264.  Back to cited text no. 1
    
2.
Ella R, Vadrevu KM, Jogdand H, Prasad S, Reddy S, Sarangi V, et al. Safety and immunogenicity of an inactivated SARS-CoV-2 vaccine, BBV152: A double-blind, randomised, phase 1 trial. Lancet Infect Dis 2021;21:637-46.  Back to cited text no. 2
    
3.
Thiagarajan K. COVID-19: India is at centre of global vaccine manufacturing, but opacity threatens public trust. BMJ 2021;372:n196.  Back to cited text no. 3
    
4.
Rawat K, Kumari P, Saha L. COVID-19 vaccine: A recent update in pipeline vaccines, their design and development strategies. Eur J Pharmacol 2021;892:173751.  Back to cited text no. 4
    
5.
Lennox RJ, Veríssimo D, Twardek WM, Davis CR, Jarić I. Sentiment analysis as a measure of conservation culture in scientific literature. Conserv Biol 2020;34:462-71.  Back to cited text no. 5
    
6.
Srivastava A, Singh V, Drall GS. Sentiment analysis of twitter data: A hybrid approach. Int J Healthc Inf Syst Inform 2019;14:1-16.  Back to cited text no. 6
    
7.
Gohil S, Vuik S, Darzi A. Sentiment analysis of health care tweets: review of the methods used. JMIR public health and surveillance. 2018;4:e5789.  Back to cited text no. 7
    
8.
Tavoschi L, Quattrone F, D'Andrea E, Ducange P, Vabanesi M, Marcelloni F, et al. Twitter as a sentinel tool to monitor public opinion on vaccination: An opinion mining analysis from September 2016 to August 2017 in Italy. Hum Vaccin Immunother 2020;16:1062-9.  Back to cited text no. 8
    
9.
Blankenship EB, Goff ME, Yin J, Tse ZT, Fu KW, Liang H, et al. Sentiment, contents, and retweets: A study of two vaccine-related twitter datasets. Perm J 2018;22:1-7.  Back to cited text no. 9
    
10.
Du J, Xu J, Song H, Liu X, Tao C. Optimization on machine learning based approaches for sentiment analysis on HPV vaccines related tweets. J Biomed Semantics 2017;8:1-7.  Back to cited text no. 10
    
11.
Dubey AD. Public Sentiment Analysis of COVID-19 Vaccination Drive in India. Available at SSRN 3772401. 2021.  Back to cited text no. 11
    
12.
Dredze M, Broniatowski DA, Smith MC, Hilyard KM. Understanding vaccine refusal: Why we need social media now. Am J Prev Med 2016;50:550-2.  Back to cited text no. 12
    
13.
Xu Z, Guo H. Using text mining to compare online pro- and anti-vaccine headlines: Word usage, sentiments, and online popularity. Commun Stud 2018;69:103-22.  Back to cited text no. 13
    
14.
Meire M, Ballings M, Van den Poel D. The added value of auxiliary data in sentiment analysis of facebook posts. Decis Support Syst 2016;89:98-112.  Back to cited text no. 14
    
15.
Salathé M, Khandelwal S. Assessing vaccination sentiments with online social media: Implications for infectious disease dynamics and control. PLoS Comput Biol 2011;7:e1002199.  Back to cited text no. 15
    


    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
 
 
    Tables

  [Table 1], [Table 2]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusions
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed310    
    Printed8    
    Emailed0    
    PDF Downloaded26    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]