Humor Detection Dataset with 10,000 English Word Expressions

该数据集包含了10,000个英文单词表达（1-2个单词），通常是词汇组合，由居住在美国的亚马逊土耳其机器人（Amazon Mechanical Turk）工作者进行标注。为了更好地区分用户偏好，所使用的单词至少拥有一个正面和一个负面注释。"二元分类幽默任务"（Binary Classification Of Humor）旨在判断这些表达是幽默的（有趣）还是不幽默（无趣）。

This dataset contains 10,000 English word expressions (1–2 words), typically lexical combinations, annotated by Amazon Mechanical Turk workers residing in the United States. To better distinguish user preferences, the selected expressions must have at least one positive and one negative annotation. The "Binary Classification of Humor" task aims to determine whether these expressions are humorous (funny) or non-humorous (unfunny).