Skip to content

Dataset · joelniklaus

Annotated Brazilian court decisions

joelniklaus/brazilian_court_decisions

4,043 judicial decision summaries from the Brazilian state of Alagoas for verdict prediction.

What's inside

The dataset contains 4,043 brief summaries (Ementa) of court decisions and metadata from the Court of Justice of Alagoas (TJAL) in Brazil. All texts are written in Portuguese and labeled into 7 categories, as well as by the unanimity of the judges' decisions. The dataset was collected for Legal Judgment Prediction tasks.

What you can do with it

  • Training legal document classification models in Portuguese
  • Building legal outcome prediction systems (LegalTech)
  • Benchmarking multilingual models in a narrow legal domain
Tasks
Text classification
Languages
Portuguese
Size
1K–10K rows
Downloads this month
367
Likes
25

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("joelniklaus/brazilian_court_decisions")