| --- |
| license: mit |
| task_categories: |
| - token-classification |
| - text-classification |
| language: |
| - en |
| tags: |
| - pii |
| - gdpr |
| - nfadp |
| - sovereign-ai |
| - switzerland |
| - australia |
| pretty_name: Cata Risk Lab Sovereign PII Markers |
| size_categories: |
| - n<1K |
| --- |
| |
| # 🛡️ Sovereign PII Detection Dataset (v1.0) |
| **Maintainer:** [Cata Risk Lab](https://huggingface.co/Cata-Risk-Lab) | **Project:** Wattle Guard |
|
|
| ## 🌍 Dataset Summary |
| This synthetic dataset contains labeled examples of **Sovereign Identity Markers** specific to the Swiss, UK, and Australian jurisdictions. It is designed to train and benchmark the **Wattle Guard** redaction engine, ensuring compliance with cross-border data protection laws (nFADP, UK GDPR, Privacy Act 1988). |
|
|
| Unlike generic PII datasets that focus on US data (SSN), this dataset focuses on international identifiers often missed by standard US-centric models. |
|
|
| ## 📂 Structure |
| - `text`: The raw input string containing potential sensitive data. |
| - `entity_type`: The specific sovereign identifier detected: |
| - `AHV13`: Swiss Social Security Number (Alters- und Hinterlassenenversicherung). |
| - `IBAN_CH`: Swiss International Bank Account Number. |
| - `NINO`: United Kingdom National Insurance Number. |
| - `NHS`: United Kingdom National Health Service Number. |
| - `TFN`: Australian Tax File Number. |
| - `label`: Binary classification (`sensitive` vs `safe`) for rapid triage. |
|
|
| ## ⚖️ Sovereign AI Mission |
| Cata Risk Lab releases this data to help open-source developers build "Region-Aware" privacy tools that respect the specific data formats of sovereign nations, rather than forcing US standards on global infrastructure.-- |
| license: mit |
| --- |
|
|
