遇见数据集

Chinese News Titles Dataset #1

收藏
arXiv2025-09-30 收录
数据链接:
官方服务:

资源简介:

该数据集包含了47,952个标题,分为32个类别用于训练,同时还有15,986个标题用于测试。为了确保比较的公正性,没有筛选出任何文本内容。该数据集的规模属于大型,任务类型为文本分类。

This dataset contains 47,952 titles, which are categorized into 32 classes for model training, while another 15,986 titles are utilized for testing. To ensure the fairness of comparative evaluations, no text content was filtered out. This is a large-scale dataset dedicated to text classification tasks.

二维码
社区交流群
二维码
科研交流群
商业服务