Humour Style Recognition Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/CrowdTruth/Short-Text-Corpus-For-Humor-Detection
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个包含1463个实例的综合性数据集，涵盖了多种幽默风格，来源多样，包括笑话和非幽默文本。具体来说，数据集中包含298个自我提升式幽默实例，265个自我贬低式幽默实例，250个亲和式幽默实例，318个攻击式幽默实例，以及332个中性实例，这样的分配确保了不同幽默风格之间的均衡代表性。该数据集的规模为1463个实例，其任务是进行幽默风格的识别。
This is a comprehensive dataset containing 1463 instances, covering various humor styles with diverse sources including jokes and non-humorous texts. Specifically, the dataset includes 298 self-improvement humor instances, 265 self-deprecating humor instances, 250 affiliative humor instances, 318 aggressive humor instances, and 332 neutral instances. Such an instance distribution ensures balanced representation across different humor styles. With a total scale of 1463 instances, the task of this dataset is humor style recognition.
提供机构：
Authors of the paper
