## Model Description
This model is trained on synthetic data generated by a Large Language Model (LLM), specifically designed for PII masking and Named Entity Recognition (NER) tasks. The dataset includes 18 well-defined entity categories spanning personal identifiers, organizational and financial details, technical attributes, temporal data, and global location information.
The training data is synthetically created, ensuring:
- No real user data is exposed.
- Balanced distribution across all entity types.
- Global coverage reflecting realistic business and enterprise use cases (e.g., banking, healthcare, technology).
- Multiple complexity levels (simple, medium, large) to simulate real-world scenarios.
### Entity Coverage
The model is trained to recognize and mask the following 18 entity types:
- Personal Identifiers: FIRSTNAME, LASTNAME, EMAIL, PHONE
- Location Information: LOCATION, ZIPCODE, COUNTRY
- Identifiers & Data: UID, NUMBER, PERCENT
- Technical: IPADDRESS, MACADD…
