If analytics tend to describe what’s happening, which is already a big step forward, the game-changing approach implements statistical models that can predict the level of the risk dynamically before, during, and, in some cases, after transactions.
These algorithms classify or score events based on a series of markers that describe their characteristics. Then, based on the results of these algorithms, i.e., a class like accept or reject or a score like a likelihood of fraud (%), a specific treatment may be defined that says how to “handle” the fraud suspicious cases.
Typically, there are three ways to handle outcomes fraud suspicious cases, i.e., to accept, verify, or reject a transaction.
|1||Accept||Automatically accepted, e.g., if it hits a whitelist.|
|2||Verify||Sent for manual verification by an analyst.|
|3||Reject||Automatically rejected, e.g., if it hits a blacklist.|