Deceptive Humor Dataset (DHD)
收藏资源简介:
Deceptive Humor Dataset(DHD)是一个新颖的资源,用于研究由虚假声明派生的幽默。该数据集由ChatGPT-4o模型生成的充满幽默感的评论组成,这些评论融入了虚假声明和操纵信息。数据集包含多种语言,包括英语、泰卢固语、印地语、卡纳达语、泰米尔语及其代码混合变体,是一个多语言基准。该数据集旨在为分析虚假语境中的幽默提供一个结构化的基础,推动一个新的研究方向,即探索幽默如何与虚假信息互动并影响其感知和传播。
Deceptive Humor Dataset (DHD) is a novel resource for researching humor derived from deceptive claims. This dataset comprises humorous comments generated by the ChatGPT-4o model, which incorporate deceptive statements and manipulated information. The dataset covers multiple languages, including English, Telugu, Hindi, Kannada, Tamil, and their code-mixed variants, serving as a multilingual benchmark. It aims to provide a structured foundation for analyzing humor within deceptive contexts, and foster a new research direction that explores how humor interacts with disinformation and influences its perception and spread.




