Deduplicated Portuguese Legal Text Corpus

LegalPT aggregates publicly available legal data in Portuguese from sources including legislation, jurisprudence, legal articles, and government documents. This version is deduplicated using the MinHash algorithm and Locality Sensitive Hashing.

Use Cases

• Train a language model on deduplicated Portuguese legal text from varied sources like legislation and jurisprudence.
• Analyze linguistic patterns across different legal corpora such as Ulysses-Tesemõ and MultiLegalPile (PT).
• Develop a retrieval system for Portuguese legal documents using the deduplicated text from government documents and legal articles.

Strengths

• Deduplicated using the MinHash algorithm and Locality Sensitive Hashing, following a published approach.
• Aggregates data from multiple legal sources including legislation, jurisprudence, and government documents.
• Composed of six distinct corpora, including Ulysses-Tesemõ and MultiLegalPile (PT).

Limitations

• Specific dataset size, row count, and column structure are unknown.
• The raw description does not specify the temporal or geographic coverage of the aggregated documents.
• The license for the dataset is not provided.

Provenance

Source

Publicly available Portuguese legal data from varied sources.

Collection Method

Aggregated and deduplicated using MinHash and Locality Sensitive Hashing.

Freshness

Last updated on 2024-05-07.

Geography

Portugal or Portuguese-speaking regions.