lightspeed-data-anonymizer

This project is derived from "data-anonymizer", developed by the UXE Data & AI
Solutions team at Red Hat
(https://gitlab.cee.redhat.com/uxe-data-ai-solutions/data-anonymizer), originally
built for the Ask Red Hat agent. The code was reused with the original authors'
permission.

Original authors / contributors to data-anonymizer:
  - Ryan Mullett (rdmullett)
  - Vamsi Mathala (vamsi-01)
  - Michael Hoang (mike-hoang)

The lightspeed-core team maintains this fork and is responsible for its tuning,
testing, accuracy, and support for Lightspeed workloads.

Accuracy note: the original authors state that redaction is not guaranteed to be
100% correct for every PII variation and that the upstream tuning targets the
Ask Red Hat workload. Validation and the confidence bar for Lightspeed workloads
are owned in this repository.

Third-party components:
  - Microsoft Presidio — https://github.com/data-privacy-stack/presidio
  - spaCy models (en_core_web_lg / en_core_web_trf) —
    https://github.com/explosion/spacy-models
  used under their respective open-source licenses.

License: Apache-2.0. See the LICENSE file.
