FakeCovid

FakeCovid是一个多语言跨领域的事实核查新闻数据集，专门针对COVID-19，由德国的杜伊斯堡-埃森大学和班贝格大学创建。该数据集包含5182篇文章，覆盖105个国家的40种语言，主要从92个不同的核查网站收集。数据集内容涉及COVID-19的多个方面，如起源、国际反应等，通过手动标注分为11个类别。创建过程中，研究者从Poynter和Snopes获取参考链接，并手动收集和标注文章。该数据集主要用于自动检测假新闻，帮助解决疫情期间的信息真实性问题。

FakeCovid is a multilingual, cross-domain fact-checking news dataset dedicated to COVID-19, created by the University of Duisburg-Essen and the University of Bamberg in Germany. This dataset contains 5,182 articles covering 40 languages from 105 countries, and was primarily collected from 92 distinct fact-checking websites. The content of the dataset covers multiple aspects of COVID-19, such as its origins and international responses, and is manually annotated into 11 categories. During its creation, researchers obtained reference links from Poynter and Snopes, and manually collected and annotated the articles. This dataset is mainly used for automatic fake news detection to help address the issue of information authenticity during the COVID-19 pandemic.