Question

Contoso's marketing team produces large volumes of unstructured internal strategy documents that do not follow a standard template and contain no consistent keywords or structured identifiers. The compliance team wants Purview to automatically identify these documents so a sensitivity label can be auto-applied, even as the writing style and topics evolve over time. Which data classification method best fits this requirement?

Accepted Answer

Create a trainable classifier and train it on sample marketing strategy documents

Answer

Configure an exact data match (EDM) sensitive info type using a schema of marketing terms

Answer

Use document fingerprinting based on a marketing strategy template

Answer

Build a custom keyword dictionary sensitive info type from common marketing phrases