Dataset Card for BBC News from FineWeb

This dataset provides a filtered subset of BBC News articles from the realnewslike subset of the FineWeb dataset, containing approximately 77k 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 crawl (from CommonCrawl between 2013-2024). 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 77k articles
No validation split in current version

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 dated subsets (i.e. not default subset nor sample 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 excluded due to sparse coverage in source data
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 which appears to be from around 2019 and earlier. For applications requiring comprehensive coverage or accurate publication dates, additional data sources should be considered.

Future Directions

Potential expansion using fineweb dataset for more recent content
Addition of publication dates through targeted crawling
Filtering to distinguish between section pages and article pages
Creation of validation split

Citation

Please cite the original FineWeb dataset when using this data. A reference to this one would be welcome but not necessary, I consider this a derivative work.

Dataset Card Authors

Louis Maddox (@permutans)