Recommendations

Important

This content is archived and is not being updated. For the latest documentation, see Microsoft Dynamics 365 product documentation. For the latest release plans, see Dynamics 365 and Microsoft Power Platform release plans.

Note

These release notes describe functionality that may not have been released yet. To see when this functionality is planned to release, please review What’s new and planned for Dynamics 365 Fraud Protection. Delivery timelines and projected functionality may change or may not ship (see Microsoft policy).

The machine learning models in the fraud protection network help find emerging fraud patterns and risky cohorts across all participating merchants. The virtual fraud analyst can leverage these findings to make recommendations to all Dynamics 365 Fraud Protection merchants about improving the configuration of their rules. To maximize the product's impact, the virtual fraud analyst also makes recommendations about how to augment existing data by adding attributes into the knowledge graph (for example, chargeback data, margins, and COGS).