Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Languages:
Portuguese
ArXiv:
Tags:
legal
Libraries:
Datasets
Dask
eduagarcia commited on
Commit
9df8fa8
·
verified ·
1 Parent(s): e71a5b3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -35
README.md CHANGED
@@ -458,6 +458,7 @@ configs:
458
  # LegalPT
459
 
460
  LegalPT aggregates the maximum amount of publicly available legal data in Portuguese, drawing from varied sources including legislation, jurisprudence, legal articles, and government documents.
 
461
 
462
  ## Dataset Details
463
 
@@ -475,46 +476,11 @@ Dataset is composed by six corpora:
475
 
476
  ### Dataset Description
477
 
478
- - **Curated by:** [More Information Needed]
479
- - **Funded by:** [More Information Needed]
480
  - **Language(s) (NLP):** Brazilian Portuguese (pt-BR)
481
  - **License:** [Creative Commons Attribution 4.0 International Public License](https://creativecommons.org/licenses/by/4.0/deed.en)
482
-
483
- ### Dataset Sources
484
-
485
  - **Repository:** https://github.com/eduagarcia/roberta-legal-portuguese
486
  - **Paper:** [More Information Needed]
487
 
488
- ## Dataset Structure
489
-
490
- <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
491
-
492
- [More Information Needed]
493
-
494
- ## Data Collection and Processing
495
-
496
- LegalPT is deduplicated using [MinHash algorithm](https://dl.acm.org/doi/abs/10.5555/647819.736184) and [Locality Sensitive Hashing](https://dspace.mit.edu/bitstream/handle/1721.1/134231/v008a014.pdf?sequence=2&isAllowed=y), following the approach of [Lee et al. (2022)](http://arxiv.org/abs/2107.06499).
497
-
498
- We used 5-grams and a signature of size 256, considering two documents to be identical if their Jaccard Similarity exceeded 0.7.
499
-
500
- Duplicate rate found by the Minhash-LSH algorithm for the LegalPT corpus:
501
-
502
- | **Corpus** | **Documents** | **Docs. after deduplication** | **Duplicates (%)** |
503
- |--------------------------|:--------------:|:-----------------------------:|:------------------:|
504
- | Ulysses-Tesemõ | 2,216,656 | 1,737,720 | 21.61 |
505
- | MultiLegalPile (PT) | | | |
506
- | CJPG | 14,068,634 | 6,260,096 | 55.50 |
507
- | BRCAD-5 | 3,128,292 | 542,680 | 82.65 |
508
- | EUR-Lex (Caselaw) | 104,312 | 78,893 | 24.37 |
509
- | EUR-Lex (Contracts) | 11,581 | 8,511 | 26.51 |
510
- | EUR-Lex (Legislation) | 232,556 | 95,024 | 59.14 |
511
- | Legal MC4 | 191,174 | 187,637 | 1.85 |
512
- | ParlamentoPT | 2,670,846 | 2,109,931 | 21.00 |
513
- | Iudicium Textum | 198,387 | 153,373 | 22.69 |
514
- | Acordãos TCU | 634,711 | 462,031 | 27.21 |
515
- | DataSTF | 737,769 | 310,119 | 57.97 |
516
- | **Total (LegalPT)** | **24,194,918** | **11,946,015** | **50.63** |
517
-
518
  ## Citation
519
 
520
  ```bibtex
 
458
  # LegalPT
459
 
460
  LegalPT aggregates the maximum amount of publicly available legal data in Portuguese, drawing from varied sources including legislation, jurisprudence, legal articles, and government documents.
461
+ This is the raw version. Deduplicated version is available [here](https://huggingface.co/datasets/eduagarcia/LegalPT_dedup).
462
 
463
  ## Dataset Details
464
 
 
476
 
477
  ### Dataset Description
478
 
 
 
479
  - **Language(s) (NLP):** Brazilian Portuguese (pt-BR)
480
  - **License:** [Creative Commons Attribution 4.0 International Public License](https://creativecommons.org/licenses/by/4.0/deed.en)
 
 
 
481
  - **Repository:** https://github.com/eduagarcia/roberta-legal-portuguese
482
  - **Paper:** [More Information Needed]
483
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
484
  ## Citation
485
 
486
  ```bibtex