Understanding covid-19 halal vaccination discourse on facebook and twitter using aspect-based sentiment analysis and text emotion analysis

Feizollah, Ali and Juma'at, Nor Badrul Anuar and Mehdi, Riyadh and Firdaus, Ahmad and Sulaiman, Ainin (2022) Understanding covid-19 halal vaccination discourse on facebook and twitter using aspect-based sentiment analysis and text emotion analysis. International Journal of Environmental Research and Public Health, 19 (10). ISSN 1660-4601, DOI https://doi.org/10.3390/ijerph19106269.

Full text not available from this repository.


The COVID-19 pandemic introduced unprecedented challenges for people and governments. Vaccines are an available solution to this pandemic. Recipients of the vaccines are of different ages, gender, and religion. Muslims follow specific Islamic guidelines that prohibit them from taking a vaccine with certain ingredients. This study aims at analyzing Facebook and Twitter data to understand the discourse related to halal vaccines using aspect-based sentiment analysis and text emotion analysis. We searched for the term ``halal vaccine'' and limited the timeline to the period between 1 January 2020, and 30 April 2021, and collected 6037 tweets and 3918 Facebook posts. We performed data preprocessing on tweets and Facebook posts and built the Latent Dirichlet Allocation (LDA) model to identify topics. Calculating the sentiment analysis for each topic was the next step. Finally, this study further investigates emotions in the data using the National Research Council of Canada Emotion Lexicon. Our analysis identified four topics in each of the Twitter dataset and Facebook dataset. Two topics of ``COVID-19 vaccine'' and ``halal vaccine'' are shared between the two datasets. The other two topics in tweets are ``halal certificate'' and ``must halal'', while ``sinovac vaccine'' and ``ulema council'' are two other topics in the Facebook dataset. The sentiment analysis shows that the sentiment toward halal vaccine is mostly neutral in Twitter data, whereas it is positive in Facebook data. The emotion analysis indicates that trust is the most present emotion among the top three emotions in both datasets, followed by anticipation and fear.

Item Type: Article
Funders: Universiti Malaya [Grant No:BKS001-2019], Impact-oriented Interdisciplinary Research Grant (IIRG) [Grant No:IIRG001C-2020SAH]
Uncontrolled Keywords: Vaccine; Halal vaccine; Covid-19; Social media; Twitter; Facebook; Sentiment analysis; Emotion analysis
Subjects: G Geography. Anthropology. Recreation > GE Environmental Sciences
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Computer Science & Information Technology > Department of Computer System & Technology
Depositing User: Ms. Juhaida Abd Rahim
Date Deposited: 11 Oct 2023 01:03
Last Modified: 11 Oct 2023 01:03
URI: http://eprints.um.edu.my/id/eprint/42228

Actions (login required)

View Item View Item