LTCR
收藏资源简介:
LTCR数据集是由中山大学智能工程学院创建,专注于长文本中文谣言检测,特别是在COVID-19相关复杂假新闻的背景下。该数据集包含2290条新闻,其中1729条为真实新闻,561条为假新闻,平均长度分别为232.5和153.5字符。数据集通过整合多种来源的数据,包括已发表论文中的数据集和未授权发布的脚本及事实核查网站,确保了数据的高质量和多样性。LTCR数据集主要用于模型训练和评估,以提高对长文本假新闻的检测能力,特别是在社交媒体平台上。
The LTCR Dataset was developed by the School of Intelligent Engineering, Sun Yat-sen University, and focuses on long-text Chinese rumor detection, specifically in the context of COVID-19-related complex fake news. This dataset comprises 2,290 news articles, with 1,729 real news items and 561 fake news items, boasting average lengths of 232.5 and 153.5 Chinese characters respectively. To ensure high data quality and diversity, it integrates data from multiple sources, including datasets from published papers, unauthorized scripts, and fact-checking websites. The LTCR Dataset is primarily intended for model training and evaluation, aiming to enhance the detection performance of long-text fake news, especially on social media platforms.




