GiusBERTo: A Legal Language Model for Personal Data De-identification in Italian Court of Auditors Decisions
Giulio Salierno, Rosamaria Bert\`e, Luca Attias, Carla Morrone, Dario, Pettazzoni, Daniela Battisti
TL;DR
GiusBERTo is a specialized BERT-based model designed for anonymizing personal data in Italian legal documents, achieving high accuracy and aiding privacy in legal data sharing.
Contribution
It introduces the first Italian legal language model for de-identification, trained on court decisions to improve privacy in legal texts.
Findings
97% token-level accuracy on test set
Effective recognition of personal data entities
Tailored for Italian legal documents
Abstract
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.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPrivacy, Security, and Data Protection · Digitalization, Law, and Regulation
MethodsRefunds@Expedia|||How do I get a full refund from Expedia? · Attention Is All You Need · Sparse Evolutionary Training · WordPiece · Residual Connection · Weight Decay · Softmax · Layer Normalization · Attention Dropout · Linear Warmup With Linear Decay