ArCovidVac: Analyzing Arabic Tweets About COVID-19 Vaccination

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 3220–3230 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 3220 ArCovidVac: Analyzing Arabic Tweets About COVID-19 Vaccination Hamdy Mubarak1, Sabit Hassan2, Shammur Absar Chowdhury1 and Firoj Alam1 1Qatar Computing Research Institute, HBKU, Qatar 2University of Pittsburgh, USA {hmubarak, shchowdhury, fialam}@hbku.edu.qa, sah259@pitt.edu Abstract The emergence of the COVID-19 pandemic and the first global infodemic have changed our lives in many different ways. We relied on social media to get the latest information about COVID-19 pandemic and at the same time to disseminate information. The content in social media consisted not only health related advise, plans, and informative news from policymakers, but also contains conspiracies and rumors. It became important to identify such information as soon as they are posted to make an actionable decision (e.g., debunking rumors, or taking certain measures for traveling). To address this challenge, we developed and publicly released the first largest manually annotated Arabic tweet dataset, ArCovidVac, for the COVID-19 vaccination campaign, covering many countries in the Arab region. The dataset is enriched with different layers of annotation, including, (i) Informativeness (more vs. less important tweets); (ii) fine-grained tweet content types (e.g., advice, rumors, restriction, authenticate news/information); and (iii) stance towards vaccination (pro-vaccination, neutral, anti-vaccination). Further, we performed in-depth analysis of the data, exploring the popularity of different vaccines, trending hashtags, topics and presence of offensiveness in the tweets. We studied the data for individual types of tweets and temporal changes in stance towards vaccine. We benchmarked the ArCovidVac dataset using transformer models for informativeness, content types, and stance detection. Keywords: COVID-19, Vaccination, Stance Detection, Arabic Tweets, Tweet Classification 1. Introduction Social media are integrated with our daily life. We share and access information through social media plat- forms making it the most prominent form of commu- nication. Due to its reach to a larger and interna- tional population, many organizations and individuals use them to circulate their contents. Thus also making these platforms a constitutive part of online news dis- tribution and consumption (Mitchell and Page, 2014). In Figure 1(a) and 1(b) we demonstrate how online users share information (e.g., rumors, plan, travel re- striction, personal experience). The post containing advice is important to reduce the spread of COVID- 19 as, for example, vaccinated people can be also a carrier. Identifying such types of content from social media become important to the government, interna- tional and local organisation for understanding psycho- logical and physical well being along with public re- actions to every taken actions. Such an understanding can (i) aid decision making by governments; and (ii) prevent rumours and fake cures that can bring harm to the society. Research studies have been conducted using numbers of COVID-19 datasets collected from Twitter. The research focused on: unlabeled (Chen et al., 2020; Banda et al., 2021; Alqurashi et al., 2020; Haouari et al., 2021a), automatically labeled (Abdul- Mageed et al., 2020; Qazi et al., 2020b), labeled us- ing distant …

Content truncated to 500 words