The minFraud services return an overall risk score, which represents the likelihood that a transaction is fraudulent. You can use the overall risk score and risk score reasons (available only through the minFraud Factors service) to determine how to respond to a transaction—whether to accept, reject, or review it more carefully.
We consider over 30 data points that can be a high signal for fraud and risk. These data points are weighted and combined with other factors such as velocity checks to compute the overall risk score. Customers cannot alter the algorithm by which these scores are combined into the overall risk score.
All of the minFraud services will return the overall risk score for your transaction. Learn more about the overall risk score.
If you pass the IP address input, all minFraud services will also return a risk factor score specific to IP risk. Learn more about the IP risk score.
In addition, the minFraud Factors service will return the risk score reasons that change the risk score significantly for a given transaction. Learn more about risk score reasons.
Customizing the risk score
The minFraud service does not allow you to natively change how the risk score is calculated. If you want to develop your own risk score, you could develop your own application that takes in risk data and risk score reasons and combines them in a way that's customized to your business. Learn more about risk data and risk score reasons.
There are a few ways that you can influence how the risk score is generated, or give you actionable outputs that are based on your customized needs. You can:
- Disable IP risk scoring for certain IP addresses. Learn how to disable IP risk scoring.
- Disable velocity checks. Velocity checks are always disabled for transactions where IP risk scoring is disabled. Learn about velocity checks and how to disable them.
- Create custom rules that return an actionable disposition of accept, reject, or review. Learn about custom rules and dispositions.
This page was last updated on .