English Wikinews Articles From ROOTS Corpus

A filtered subset of English-language Wikinews articles from the ROOTS corpus. It represents 0.0271% of the total English content within the larger ROOTS collection. The data was processed and released by BigScience in December 2022.

Use Cases

• Analyze text patterns and journalistic style in English Wikinews articles from the ROOTS corpus.
• Train language models on a curated subset of news text filtered from the larger ROOTS dataset.
• Study the characteristics of content that passed the 'filter_wiki_user_titles' and 'filter_wiki_non_text_type' preprocessing steps.

Strengths

• Data is part of the large-scale, multilingual ROOTS corpus curated by BigScience.
• Specific preprocessing filters like 'filter_wiki_user_titles' were applied to ensure data quality.
• Released in a structured format with metadata tags indicating text modality and a CC BY-SA 3.0 license.

Limitations

• The dataset size is unspecified, with no row count, column details, or sample data available for assessment.
• The sample represents only a tiny fraction (0.0271%) of the English content in ROOTS, which may limit its scope.
• The last update was in December 2022, so the news content is not current.

Provenance

Source

Wikinews, processed as part of the BigScience ROOTS corpus.

Collection Method

Filtered from the larger ROOTS dataset using steps including 'filter_wiki_user_titles' and 'filter_wiki_non_text_type'.

Freshness

Last updated on 2022-12-12.

Geography

Region tag indicates 'us', but specific geographic coverage is unknown.