Sentimental LIAR
收藏arXiv2020-10-22 更新2024-06-21 收录
下载链接：
https://github.com/UNHSAILLab/SentimentalLIAR
下载链接
链接失效反馈官方服务：
资源简介：
Sentimental LIAR是由新罕布什尔大学SAIL实验室开发的扩展数据集，基于原始的LIAR数据集，增加了基于情感和情绪分析的特征。该数据集包含12,836条短声明，每条声明都标注了真假，并附有情感和情绪分析结果。创建过程中，研究者使用了Google和IBM的NLP API来分析情感和情绪。该数据集主要应用于社交媒体上的假声明检测，旨在通过自动化技术提高信息的真实性和可信度。
Sentimental LIAR is an extended dataset developed by the SAIL Lab at the University of New Hampshire, built upon the original LIAR dataset and augmented with features derived from sentiment and emotion analysis. This dataset contains 12,836 short statements, each annotated with its truthfulness label along with corresponding sentiment and emotion analysis results. During the dataset construction, researchers employed NLP APIs from Google and IBM to conduct sentiment and emotion analyses. This dataset is primarily utilized for false statement detection on social media, aiming to improve information authenticity and credibility through automated techniques.
提供机构：
新罕布什尔大学SAIL实验室
创建时间：
2020-09-01
