ABSTRACT
Vaccines are efficient invented weapons to save millions of lives during the global pandemics such as COVID19. The public attitudes to COVID-19 vaccination process should be studied carefully in order to understand the society's opinions and concerns toward it, and to take the appropriate plans by the public health authorities accordingly. The availability of accurate information during the pandemic, its reported sources and their confidentiality play a key role on the public reported opinions toward the vaccination topic. The society used the social media extensively during the pandemic and quarantine periods to track the COVID-19 news and follow the updates of the vaccines' development process. In this study, we focused on revealing the public attitudes of the COVID-19 vaccination process on twitter as a social media platform using machine learning. The tweets are collected and preprocessed for features extraction and reduction step and the K-means clustering algorithm is used to group the vaccines related tweets in order to use the Amazon Comprehend module for sentiment analysis. Most of the reported opinions that discussed the vaccines efficiency, safety, and the governments distribution plans and the policies to secure the doses for their residents were neutral. The second largest group has negative opinions according to the inaccurate information about vaccines development process, their side effects, and unsuccessful reported trials of vaccines production during different pandemic periods. Our study highlighted the urgent need for interactive communications with the society from different cultural and educational backgrounds in order to increase the vaccination awareness and validate its related news. The deliverable speech of the health care decision makers should simplify the vaccines related scientific terms, address the community concerns, and make the vaccination distribution plans publicly available for their underlying communities. © 2021 IEEE.
ABSTRACT
Since the declaration of COVID-19 as a global pandemic, the world is disrupting socially and economically. COVID-19 vaccines are still in the human trials and the governments should understand the society attitudes towards the vaccinations acceptance/hesitancy in order to deliver more accurate plans and health messages. Social media represents a catalog of our daily-life communications and activities. In this paper, we utilized the power of social media and machine learning to gain insights and understand the public attitudes towards COVID-19 vaccinations. The peak of online vaccination conversation on a social media happened with the authorization of Moderna and Pfizer vaccines for the emergency usage. The sentiment analysis of the clustered tweets relevant to the vaccination topic reveals that most of the public opinions was neutral and target the understanding of the vaccination process, confirming its efficiency and safety, and the countries plans to distribute and secure doses for their residents. The second top sentiment analysis group has negative attitudes according to spreading claims of the vaccines productions, side effects, and the previously reported vaccinations trails in different historical pandemic periods. The reported analysis raised the emergency need for interactive communications with communities from different cultural and educational level to increase their vaccination awareness and validate the vaccines' associated news. The decision makers' deliverable speech should simplify the scientific terms, target the community fears and release any public concerns regarding the vaccination process and its distribution plans. © 2021 Association for Computing Machinery.