amyguan/newswire-20-30

该数据集包含多个特征，如文章内容、作者、日期、报纸元数据、多个主题标签（如反垄断、民权、犯罪等）、命名实体识别（NER）相关的词汇和标签、电线信息、提及的人物信息等。数据集主要用于训练，包含361,779个样本，总大小为1,803,139,033字节。

This dataset contains news articles along with their associated metadata, such as article content, author information, dates, newspaper metadata (including newspaper title, city, state, etc.), and topic labels related to the articles (e.g., antitrust, civil rights, crime, etc.). Additionally, it includes Named Entity Recognition (NER) information, geolocation data, and information about people mentioned in the articles (such as gender, occupation, Wikidata ID, etc.). The dataset is divided into a training set (train) with 361779 samples.