Haniehedi/ag_news_annotated

该数据集名为ag_news_annotated，是基于AG News数据集创建的注释数据集，使用Argilla工具进行构建和管理。它主要用于文本分类和实体标注任务，包含新闻文本字段（text），以及两个注释问题：分类标签（label，可选值为Business、Sci/Tech、Sports、World）和实体标注（entities，用于高亮文本中的实体）。数据集结构包括单个训练分割（train），示例数据展示了新闻文本及其分类建议。数据集旨在支持自然语言处理研究，特别是新闻分类和实体识别应用。This dataset, named ag_news_annotated, is an annotated dataset built upon the AG News benchmark dataset, and was constructed and managed using the Argilla tool. It is primarily designed for text classification and entity annotation tasks, containing a news text field (`text`), along with two annotation tasks: a classification label (`label`) with optional values of Business, Sci/Tech, Sports, and World, and entity annotations (`entities`) which are used to highlight entities within the text. The dataset structure includes a single training split (`train`), with example data showcasing news texts and their proposed classification labels. This dataset aims to support natural language processing (NLP) research, particularly for news classification and entity recognition applications.