Portuguese Legal Sentences from the Supreme Court of Justice, Extended Version

A collection of 500,000 legal sentences from the Portuguese Supreme Court of Justice, created by rufimelo. The dataset was specifically built for use in Masked Language Modeling (MLM) and Transformer-based Sequential Denoising Auto-Encoder (TSDAE) tasks. It is an extended version of the PortugueseLegalSentences-v1 dataset and was last updated on November 1, -2022.

Use Cases

• Training Masked Language Models (MLM) based on Portuguese legal text.
• Fine-tuning sentence embedding models using TSDAE based on the legal sentence collection.
• Benchmarking domain adaptation techniques for Portuguese NLP.
• Analyzing linguistic patterns and terminology within Portuguese Supreme Court rulings.

Strengths

• Contains 500,000 sentences, providing substantial text volume for training.
• Specifically designed for MLM and TSDAE tasks, indicating a clear purpose.
• Sourced from the Portuguese Supreme Court of Justice, offering authoritative legal text.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last updated 2022-11-01 13:14:38; freshness should be verified.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Portuguese Supreme Court of Justice.

Collection Method

Collection of legal sentences; specific gathering method is not detailed.

Time Range

null

Freshness

Last updated 2022-11-01 13:14:38.

Geography

Portugal.