Rumors about COVID-19 on Weibo

In order to systematically study the characteristics of the topological structure of the true and false news of COVID-19, we used the Weibo API to collect and construct a new data set with a time span from November 2019 to March 2022, including 4184 source posts and a total of 961,975 microblogs with their retweets and comments. The topology of the dissemination cascade owned by each source microblog is completely recorded, and the user profile of the author of each microblog is also collected. The false news comes from Weibo Community Management Center, an official service where users can report either a microblog that contains false information, and the true news comes from official accounts, the rumor-refuting microblogs provided by Weibo Community Management Center and China Internet Joint Rumor Debunking Platform. Our dataset contains 1407 pieces of true information and 768 pieces of misinformation about COVID-19. We encode the content of microblogs in the dataset using TF-IDF with top-5000 words. For more guidance on using the dataset, please refer to https://github.com/kkkkk001/KPG/tree/main

搜集汇总

背景与挑战

背景概述

该数据集是一个专门针对COVID-19在微博平台上谣言传播的研究数据集，覆盖2019年11月至2022年3月的时间范围，包含4184条源帖和总计961,975条微博，并完整记录了传播拓扑结构和用户信息。数据集标注了1407条真实信息和768条虚假信息，采用TF-IDF编码，适用于计算机与信息科学领域的谣言检测和分析研究。