Hugging Face's logo

Datasets:
QomSSLab
/
legal_full_v4

Tasks:
Text Classification
Modalities:
Tabular
Text
Formats:
parquet
Languages:
Persian
Size:
100K - 1M
Libraries:
Datasets
Dask
License:
Dataset card Data Studio Files Community
1

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

legal_full_v4

legal_full_v4 is a curated subset of the legal_full_v3 dataset, filtered to exclude laws passed before the 1979 Iranian revolution (before the year 1357 in the Persian calendar) and also 'هیات وزیران' in row['tags'].

Original Dataset Description (legal_full_v3)

The original legal_full_v3 dataset was created by crawling and parsing the structure of the website containing law documents.

Downloads last month
140
Source
.
Filtering
: All legal documents dated before 1357 have been removed using regex-based rules on date fields to provide a post-revolution legal corpus.
Use case
: Suitable for legal text classification, retrieval, and NLP modeling focused on modern Iranian laws.