metadata

dataset_info:
  - config_name: CC-MAIN-2013-20
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 554836050
        num_examples: 179829
    download_size: 342273044
    dataset_size: 554836050
  - config_name: CC-MAIN-2013-48
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 1201156640
        num_examples: 394978
    download_size: 742585587
    dataset_size: 1201156640
  - config_name: CC-MAIN-2014-10
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 1224875616
        num_examples: 405967
    download_size: 757458265
    dataset_size: 1224875616
  - config_name: CC-MAIN-2014-15
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 899256194
        num_examples: 295272
    download_size: 555731673
    dataset_size: 899256194
  - config_name: CC-MAIN-2014-23
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 1222018466
        num_examples: 409959
    download_size: 755291616
    dataset_size: 1222018466
  - config_name: CC-MAIN-2014-35
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 1224453287
        num_examples: 410024
    download_size: 757204790
    dataset_size: 1224453287
  - config_name: sample-10BT
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 29594151
        num_examples: 9884
    download_size: 18373460
    dataset_size: 29594151
  - config_name: sample-350BT
    features:
      - name: url
        dtype: string
      - name: text
        dtype: string
    splits:
      - name: train
        num_bytes: 1041478873
        num_examples: 352549
    download_size: 646136439
    dataset_size: 1041478873
configs:
  - config_name: CC-MAIN-2013-20
    data_files:
      - split: train
        path: CC-MAIN-2013-20/train-*
  - config_name: CC-MAIN-2013-48
    data_files:
      - split: train
        path: CC-MAIN-2013-48/train-*
  - config_name: CC-MAIN-2014-10
    data_files:
      - split: train
        path: CC-MAIN-2014-10/train-*
  - config_name: CC-MAIN-2014-15
    data_files:
      - split: train
        path: CC-MAIN-2014-15/train-*
  - config_name: CC-MAIN-2014-23
    data_files:
      - split: train
        path: CC-MAIN-2014-23/train-*
  - config_name: CC-MAIN-2014-35
    data_files:
      - split: train
        path: CC-MAIN-2014-35/train-*
  - config_name: sample-10BT
    data_files:
      - split: train
        path: sample-10BT/train-*
  - config_name: sample-350BT
    data_files:
      - split: train
        path: sample-350BT/train-*
license: odc-by
language:
  - en
tags:
  - news
  - fineweb
pretty_name: FineWeb BBC News
size_categories:
  - 1M<n<10M

Dataset Card for BBC News from FineWeb

This dataset provides a filtered subset of BBC News articles from each subset of the FineWeb dataset, expected to contain approximately 300M articles from BBC News domains.

Dataset Details

Dataset Description

Curated by: Louis Maddox (@permutans on HuggingFace and X/Twitter)
License: ODC-BY (inherited from FineWeb)
Language: English

Dataset Sources

Repository: https://huggingface.co/datasets/permutans/fineweb-bbc-news
Source Dataset: HuggingFaceFW/fineweb
Paper: https://arxiv.org/abs/2406.17557 (FineWeb paper)

Uses

Direct Use

Suitable for text analysis and NLP tasks focused on news content, particularly when working with BBC News articles. The dataset provides cleaned article text without metadata like bylines or publication dates.

Out-of-Scope Use

This dataset should not be used as a comprehensive archive of BBC News content, as it represents only articles captured in FineWeb's crawls. It should not be assumed to contain all articles from any given time period.

Dataset Structure

Data Instances

Example format:

{
    'url': 'news.bbc.co.uk/news/article-path',
    'text': 'Article content...'
}

Data Fields

url: URL of the article with query parameters removed
text: Full article text content

Data Statistics

Contains approximately 300k articles per subset

Dataset Creation

Curation Rationale

Created to provide an easily accessible dataset of BBC news articles while offering a focused view into the FineWeb dataset's coverage of major news sources. Enables analysis of FineWeb's completeness and motivates investigation of alternative data acquisition methods.

Source Data

Data Collection and Processing

Filtered from FineWeb's subsets
Limited to domains: news.bbc.co.uk, www.bbc.co.uk/news, www.bbc.com/news
URL cleaning: removed query parameters
Regional news content may be excluded (under regional subsite paths rather than /news)
No modifications to article text content

Personal and Sensitive Information

Article texts contain only the main content body, without bylines or metadata.

Bias, Risks, and Limitations

No validation split in current version
Original publication dates not available (FineWeb timestamps were crawl dates)
Section/index pages not yet filtered out from article pages
Regional news content explicitly excluded due to sparse coverage
Relationship between news.bbc.co.uk and bbc.co.uk/news domains needs investigation
Coverage may be incomplete compared to full BBC News archive

Recommendations

Users should be aware that this represents a subset of BBC News content. For applications requiring comprehensive coverage or accurate publication dates, additional data sources should be considered.

Future Directions

Addition of publication dates through targeted crawling
Filtering to distinguish between section pages and article pages

Citation

Please cite the original FineWeb dataset when using this data.

Dataset Card Authors

Louis Maddox (@permutans)