README.md

metadata

license: mit
language:
  - zh
task_categories:
  - text-classification
  - question-answering
  - summarization
tags:
  - caexpo
  - china-asean
  - news
pretty_name: CAEXPO News Dataset
size_categories:
  - 1K<n<10K

CAEXPO News Dataset

Dataset Description

The CAEXPO (China-ASEAN Expo) News Dataset is a comprehensive collection of news articles from the offical China-ASEAN Expo website. The dataset covers various aspects including:

• Trade and economic cooperation
• Cultural exchanges
• Regional development
• Policy announcements
• Innovation and technology cooperation
• ......

Data Fields

• title: The headline of the news article
• url: URL link to the original news article
• publishTime: Publication timestamp of the article
• description: Brief summary of the article(may be empty)
• sourceNew: Source of the news article(may be empty)
• markdown: Full content of the article in markdown format(Converted from HTML to Markdown by the deepseek-chat model.)

Dataset Structure

Data Instances

Example of a typical data instance:

{
    "title": "中国—东盟青年领军者成长计划大咖沙龙——与专家面对面活动成功举办",
    "url": "https://www.caexpo.org.cn/zh-CN/posts/831448314061901824",
    "publishTime": "2024-12-12T08:34:19",
    "description": "2024年12月11日，中国—东盟青年领军者成长计划大咖沙龙——与专家面对面活动在线上举办。",
    "sourceNew": "中国—东盟博览会官方微信公众号",
    "markdown": "2024年12月11日，中国—东盟青年领军者成长计划大咖沙龙——与专家面对面活动在线上举办..."
}

Usage

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("GXMZU/caexpo_news")

# Print the first example
print("Example news item:")
print(f"Title: {dataset['train'][0]['title']}")

Limitations

• The primary language of the dataset is Chinese
• Some articles may have empty description or source fields

License

This dataset is licensed under MIT.

Citation

If you use this dataset in your research, please cite:

@misc{caexpo_news_dataset,
  title={CAEXPO News Dataset},
  author={real-jiakai},
  year={2025},
  url={https://huggingface.co/datasets/GXMZU/caexpo_news/}
}