Uzbek Text Classification Dataset

本研究介绍了一个新的乌兹别克语文本分类数据集，名为Uzbek Text Classification Dataset。该数据集由Urgench State University创建，包含从10个不同新闻和媒体网站收集的超过512,750篇新闻文章，涵盖15个新闻类别，总计超过120M字。数据集创建过程中，通过网络爬虫技术收集数据，并进行了去重、文本清理和分类标签的调整。该数据集主要用于乌兹别克语的自然语言处理研究，特别是文本分类任务，旨在提高对低资源语言处理的能力。

This study introduces a novel Uzbek text classification dataset named Uzbek Text Classification Dataset. Developed by Urgench State University, this dataset comprises over 512,750 news articles collected from 10 distinct news and media websites, covering 15 news categories with a total corpus size exceeding 120 million words. During the dataset construction process, data was gathered using web crawling technologies, followed by deduplication, text cleaning, and refinement of classification labels. This dataset is primarily intended for Uzbek natural language processing research, especially text classification tasks, with the goal of advancing the processing capabilities of low-resource languages.