Publication Date


Document Type


Committee Members

William Romine, Ph.D. (Advisor); Jeffrey Peters, Ph.D. (Committee Member); Paula Bubulya, Ph.D. (Committee Member)

Degree Name

Master of Science (MS)


This study assesses the underlying topics, sentiment, and types of information regarding COVID-19 vaccines on Twitter during the initiation of the vaccine rollout. Tweets about the COVID-19 vaccine were collected and the relevant tweets were then filtered out using a relevancy classifier. Latent Dirichlet Allocation (LDA) was used to uncover topics of discussion within the relevant tweets. The NRC lexicon was used to assess positive and negative sentiment within tweets. The type of information (information, misinformation, opinion, or question) in tweets was evaluated. The relevancy classifier resulted in a dataset of 210,657 relevant tweets. Eight topics provided the best representation of the relevant tweets. Tweets with negative sentiment were associated with a higher percentage of misinformation. Tweets with positive sentiment showed a higher percentage of information. The proliferation of information and misinformation on social media platforms are associated with building trust and mitigating negative sentiment associated with COVID-19 vaccines.

Page Count


Department or Program

Department of Biological Sciences

Year Degree Awarded


Included in

Biology Commons