Living Legal Dataset for Text Classification

Living legal dataset created by pile-of-law, last updated on July 7, 2022. The dataset is monolingual in English, sourced from the US region, and is tagged for text classification tasks.

Use Cases

• Train a text classification model on legal documents using the 'Task Categories:text Classification' tag as a guide.
• Analyze the linguistic patterns of US legal texts, leveraging the 'Language:en' and 'Region:us' metadata.
• Use the dataset's 'Multilinguality:monolingual' nature to build or benchmark models focused on English legal language processing.

Strengths

• Dataset is described as 'living', suggesting ongoing updates and maintenance.
• Specific metadata tags provide clear context for the data's language, region, and intended task.

Limitations

• Critical descriptive information such as row count, column structure, and sample data is unavailable.
• The dataset's size, file formats, and specific license details are unknown, hindering usability assessment.

Provenance

Source

pile-of-law

Freshness

Last updated on 2022-07-07.

Geography

United States (based on 'Region:us' tag)