Skip to content

Dataset · vblagoje

English CC-News articles

vblagoje/cc_news

708k English news articles from 2017–2019 sourced from Common Crawl.

What's inside

The corpus contains 708,241 English news articles published from January 2017 to December 2019. The data was prepared using news-please and retrieved from Common Crawl (CC-NEWS bucket). Image and text modalities are specified; this is a small subset of the full CC-News.

What you can do with it

  • Fine-tune a language model on news texts
  • Evaluate summarization or entity extraction from news
  • Use for English fill-mask tasks
Tasks
Text generation, Fill mask
Languages
English
Size
100K–1M rows
Downloads this month
5.9K
Likes
70

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("vblagoje/cc_news")