COVID-19 Fake News Dataset
收藏arXiv2025-09-30 收录
下载链接:
https://competitions.codalab.org/competitions/26655
下载链接
链接失效反馈官方服务:
资源简介:
该数据集来源于多个社交媒体和事实核查网站,专注于收集有关COVID-19新闻的项目。它包含了经核实来源的真实新闻和虚假新闻文章,并包含了提取统计特征的属性。在类别平衡方面,真实新闻占52.34%,虚假新闻占47.66%。数据集中包含了880个独特的用户名和210个独特的URL域名。规模上,该数据集包含了10,700条社交媒体新闻条目,其任务是进行虚假新闻检测。
This dataset is sourced from multiple social media platforms and fact-checking websites, focusing on collecting COVID-19 news-related content. It includes real and fake news articles with verified sources, alongside attributes for extracting statistical features. In terms of class balance, real news accounts for 52.34% of the dataset, while fake news makes up 47.66%. The dataset contains 880 unique usernames and 210 unique URL domains. With a total of 10,700 social media news entries, its target task is fake news detection.
提供机构:
Constraint COVID-19 Fake News Detection Challenge organizers



