Congressional Speeches
收藏arXiv2025-09-30 收录
下载链接：
https://huggingface.co/datasets/hazylavender/CongressionalDataset
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了从美国、英国和加拿大的国会或议会发言记录中搜集的13.4万场演讲或辩论，每个演讲都被视为独立的客户端。数据样本是通过对演讲内容进行连续的64个标记划分来创建的。此外，该数据集每六个月更新一次，规模达到了13.4万场演讲，所涉及的任务是联邦学习。
This dataset includes 134,000 speeches and debates collected from congressional or parliamentary records across the United States, the United Kingdom, and Canada, where each individual speech is treated as an independent client. Data samples are generated by segmenting speech content into consecutive 64-token chunks. Additionally, this dataset is updated every six months, and the underlying associated task is federated learning.
提供机构：
Hugging Face
