U.S. Parliamentary Speech Corpus Processed for NLP

En Corpora Parliament Processed is a text dataset containing parliamentary proceedings, likely from the U.S. region. The dataset was created by author JonathanSum and last updated in February 2022. Its size is categorized as between 1 million and 10 million entries.

Use Cases

• Train language models on formal political speech using the processed text content.
• Analyze topic trends or sentiment over time within parliamentary speeches.
• Perform named entity recognition to identify mentioned politicians, organizations, or legislation.
• Study linguistic features and rhetorical styles specific to government proceedings.

Strengths

• Dataset size is categorized as 1M to 10M entries, indicating a substantial text corpus.
• Data is provided in Parquet format, optimized for efficient storage and querying.

Limitations

• Specific column names and the exact number of rows are unknown, hindering precise assessment.
• The source and temporal range of the original parliamentary data are not specified.
• Data is from 2022 and may not include recent proceedings.

Provenance

Source

Original parliamentary proceedings source unspecified.

Collection Method

Processing method for the raw text is not described.

Freshness

Last updated 2022-02-22; no update frequency stated.

Geography

Platform tags suggest a 'us' region focus.