Trainable classifier auto-labeling with sensitivity labels preview

As part of this preview, the Microsoft 365 compliance center will allow you to create sensitivity labels and corresponding automatic or recommended labeling policies in Office apps using built-in classifiers.

The six built-in classifiers that are available as part of this preview are:

• Resume: detects written accounts of an applicant's personal, educational, and professional qualifications and experience
• Source code: detects a set of instructions and statements written in the top 25 computer programming languages of GitHub
• Offensive language: detects text items that contain profanities, slurs, taunts, and disguised expressions (expressions that have the same meaning as more offensive terms)
• Threat: detects a specific category of offensive language related to threat to commit violence or do physical harm/damage to a person/property.
• Harassment: detects a specific category of offensive language related to offensive conduct targeting one or multiple individuals regarding race, color, religion, national origin, gender, sexual orientation, age, disability and genetic information.
• Profanity: detects a specific category of offensive language that contains swear words or vulgar language.

References:

• Preview Blog announcement
• Getting started with trainable classifiers
• Join this private preview program at: https://aka.ms/MLSensitivityLabelsPreviewConsent

Thanks to those of you who participated in our previews so far. If you haven’t already, don’t forget to check out our preview programs page and our resources available on the Tech Community.

Thanks!

Written by @Nir Hendler , posted by @Adam Bell  on behalf of the MIP and Compliance CXE team