(Enhancement) Additional use cases for proofreading: Personal information, Abusive language detection #6
Description
Opening an issue to share some other use cases that this API could be extended for, to see if there is interest in this space.
- Checking for unintended personal information in posts.
Would it be useful for the proofreader API to support identifying if a piece of text has personal information that was shared unintentionally. For example, say a web app wants to prevent child accounts from posting phone numbers, addresses without consent.
- Checking for abusive language.
Perhaps the underlying primitive is a text analysis API.
Given a piece of text and a set of labels return [{label, sections of text matching the label, suggested fix, explanation}].