Exam SC-400 topic 2 question 46 discussion

Built-in Trainable classifiers Microsoft 365 comes with five classifiers already trained and ready to use. • Résumés • Source Code • Harassment • Profanity • Threat Enjoy, with love

Trainable classifier

Just tested: Trainable classifier > Source code

Trainable classifier

Trainable classifier is a better choise.

The sensitive info types specify which specific type, neither of them say "source code sensitive info type", so neither listed would work.

It says "The solution must minimize administrative effort." So Option D is minimize administrative effort. If you use SIT for source code, wouldn't you perform more effort?

Option C

It should be B

There is already a Sensitive Info Type for source code. So it should not be trainable classifier

There is a source code trainable classifier

I owuld say D. a source code trainable classifier is pretrained in PurView. Less effort than monitor source code dictionary

Sensitive info types in SharePoint Online help you identify and protect sensitive content such as credit card numbers, social security numbers, and source code. In this case, you want to prevent the sharing of files that contain source code. A sensitive info type that uses a keyword dictionary allows you to define a list of specific keywords or phrases associated with source code. SharePoint Online will scan the content and metadata of files for matches against the keyword dictionary, and if a match is found, it can trigger appropriate actions such as blocking or alerting. This approach minimizes administrative effort because you can maintain and update the keyword dictionary as needed without requiring extensive manual configuration or training. Therefore, option B is the correct choice for creating a DLP policy to prevent the sharing of files containing source code in Microsoft SharePoint Online.

Just use the trainable classifier that have been configured and trained already to look for source code

Isn't it an administrative effort creating a trainable classifier here?