A Dataset of State-Censored Tweets

本数据集名为'A Dataset of State-Censored Tweets'，由洛桑联邦理工学院创建，收录了2012年至2020年7月间被政府审查的583,437条推文，涉及155,715名用户。数据集通过互联网档案馆的Twitter流抓取技术收集，不仅包含被审查的推文，还包括相关用户的其他推文及转发。此数据集旨在研究政府对社交媒体的审查政策及其对用户行为的影响，同时也用于研究仇恨言论检测和审查对社交媒体用户的影响。

This dataset is named "A Dataset of State-Censored Tweets", and was created by École Polytechnique Fédérale de Lausanne. It contains 583,437 government-censored tweets published between 2012 and July 2020, involving 155,715 distinct users. The dataset was collected using Twitter stream scraping technologies from the Internet Archive, and includes not only the censored tweets themselves but also other tweets and retweets posted by the corresponding users. This dataset is designed for research on government social media censorship policies and their impacts on user behaviors, as well as studies on hate speech detection and the effects of censorship on social media users.