NVIDIA-NeMo / NVIDIA-NeMo/Anonymizer

Add entity sensitivity guidance

Open
#145 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
122
Forks
17
Avg merge
4d 2h
Merged PRs (30d)
11

Description

Priority Level

Medium (Nice to have)

Is your feature request related to a problem?

In the sensitivity disposition we assign each entity a sensitivity value, but give no guidance on how to assign it. We've seen some questionable assignments. A leaked high-sensitivity entity contributes 1.0 to leakage mass; a low one only 0.3. Also it triggers mandatory repair regardless of total leakage mass (when repair_any_high_leak=True in the risk tolerance bundle). This ticket is to add more guidance on how this value is assigned.

Describe the solution you'd like

Update the sensitivity disposition to give more guidance. This will take some thought. Certainly direct identifiers (names, SSN, phone number) are all high. Quasi-identifier sensitivity should probably depend on how much they add to re-identification risk. Should all "leave-as-is" dispositions be low sensitivity?

Describe alternatives you've considered

No response

Additional context

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the sensitivity disposition and the risk tolerance bundle, especially the repair_any_high_leak behavior described here. Define guidance for direct identifiers, quasi-identifiers, and leave-as-is dispositions, then document how assignments affect leakage mass and mandatory repair. Done means the guidance resolves the listed assignment questions without changing the stated behavior.

Written by the indexing model from the issue text.

Assessment

Domain
documentation
Issue type
Documentation
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.