USClassActions is an English dataset of 200 complaints from the US Federal Court with the respective binarized judgment outcome (Win/Lose). The dataset poses a challenging text classification task. We are happy to share this dataset in order to promote robustness and fairness studies on the critical area of legal NLP. The data was annotated using Darrow.ai proprietary tool.
from datasets import load_dataset
dataset =… See the full description on the dataset page:
https://huggingface.co/datasets/darrow-ai/USClassActionOutcomes_ExpertsAnnotations.
