COVID-19 Vaccine Hesitancy on Social Media: Building a Public Twitter Dataset of Anti-vaccine Content, Vaccine Misinformation and Conspiracies

False claims about COVID-19 vaccines can undermine public trust in ongoing\nvaccination campaigns, thus posing a threat to global public health.\nMisinformation originating from various sources has been spreading online since\nthe beginning of the COVID-19 pandemic. In this paper, we present a dataset of\nTwitter posts that exhibit a strong anti-vaccine stance. The dataset consists\nof two parts: a) a streaming keyword-centered data collection with more than\n1.8 million tweets, and b) a historical account-level collection with more than\n135 million tweets. The former leverages the Twitter streaming API to follow a\nset of specific vaccine-related keywords starting from mid-October 2020. The\nlatter consists of all historical tweets of 70K accounts that were engaged in\nthe active spreading of anti-vaccine narratives. We present descriptive\nanalyses showing the volume of activity over time, geographical distributions,\ntopics, news sources, and inferred account political leaning. This dataset can\nbe used in studying anti-vaccine misinformation on social media and enable a\nbetter understanding of vaccine hesitancy. In compliance with Twitter's Terms\nof Service, our anonymized dataset is publicly available at:\nhttps://github.com/gmuric/avax-tweets-dataset\n