GiusBERTo: A Legal Language Model for Personal Data De-identification in Italian Court of Auditors Decisions
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929394089459712 |
|---|---|
| author | Salierno, Giulio Bertè, Rosamaria Attias, Luca Morrone, Carla Pettazzoni, Dario Battisti, Daniela |
| author_facet | Salierno, Giulio Bertè, Rosamaria Attias, Luca Morrone, Carla Pettazzoni, Dario Battisti, Daniela |
| contents | Recent advances in Natural Language Processing have demonstrated the effectiveness of pretrained language models like BERT for a variety of downstream tasks. We present GiusBERTo, the first BERT-based model specialized for anonymizing personal data in Italian legal documents. GiusBERTo is trained on a large dataset of Court of Auditors decisions to recognize entities to anonymize, including names, dates, locations, while retaining contextual relevance. We evaluate GiusBERTo on a held-out test set and achieve 97% token-level accuracy. GiusBERTo provides the Italian legal community with an accurate and tailored BERT model for de-identification, balancing privacy and data protection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_15032 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GiusBERTo: A Legal Language Model for Personal Data De-identification in Italian Court of Auditors Decisions Salierno, Giulio Bertè, Rosamaria Attias, Luca Morrone, Carla Pettazzoni, Dario Battisti, Daniela Computation and Language Artificial Intelligence Recent advances in Natural Language Processing have demonstrated the effectiveness of pretrained language models like BERT for a variety of downstream tasks. We present GiusBERTo, the first BERT-based model specialized for anonymizing personal data in Italian legal documents. GiusBERTo is trained on a large dataset of Court of Auditors decisions to recognize entities to anonymize, including names, dates, locations, while retaining contextual relevance. We evaluate GiusBERTo on a held-out test set and achieve 97% token-level accuracy. GiusBERTo provides the Italian legal community with an accurate and tailored BERT model for de-identification, balancing privacy and data protection. |
| title | GiusBERTo: A Legal Language Model for Personal Data De-identification in Italian Court of Auditors Decisions |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2406.15032 |