WICO Text: A Labeled Dataset of Conspiracy Theory and 5G-Corona Misinformation Tweets

The COVID-19 pandemic has been accompanied by a flood of mis-information on social media, which has been labeled an “infodemic”.While a large part of such fake news is ultimately inconsequential,some of it has the potential to real-world harm, but due to themassive amount of social media contents, it is impossible to findthis misinformation manually. Thus, conventional fact-checkingcan typically only counteract misinformation narratives after theyhave gained significant traction. Only automated systems can pro-vide warnings in advance. However, the automatic detection ofmisinformation narratives is very challenging since the texts thatspread misinformation may be short messages on Twitter. Theymay also transmit misinformation by implication rather than bystating counterfactual information outright, and satirical messagescomplicate the issue further. Thus, there is a need for highly sophis-ticated detection systems. In order to support their development,we created substantial ground truth data by human annotation. Inthis paper, we present a dataset that deals with a specific piece ofmisinformation: the idea that the COVID-19 pandemic is causallyconnected to the 5G wireless network. We selected more than 10,000tweets that deal with COVID-19 and 5G and labeled them manually,distinguishing between tweets that propagate the specific 5G misin-formation, those that spread other conspiracy theories, and tweetsthat do neither. We provide the human-annotated dataset alongwith an additional large-scale automatically (by using the human-annotated dataset as the training set) labelled dataset consist ofmore than 100,000 tweets

Download paper here

Bibtex

@inproceedings{pogorelov2021wico,
  title={Wico text: a labeled dataset of conspiracy theory and 5g-corona misinformation tweets},
  author={Pogorelov, Konstantin and Schroeder, Daniel Thilo and Filkukov{\'a}, Petra and Brenner, Stefan and Langguth, Johannes},
  booktitle={Proceedings of the 2021 Workshop on Open Challenges in Online Social Networks},
  pages={21--25},
  year={2021}
}