In this section, the color scheme shown in the screenshots may differ from the actual appearance of the Control Centre.
On this page, you will find details about the Fraud Rules module, that allows you to set up the needed traffic protection policy with the use of the AI-based methods: Proactive Monitoring and Behavioral Analysis.
We strongly advise to read the Recommendations section before making any changes in this module.
ΒΆ How to Set up Fraud Rules
Our AI engine, used in Proactive Monitoring and Behavioral Analysis methods, allows the system to detect very well-disguised fraud attacks and even fraud preparations on hubbing and retail traffic.
These methods utilize unique approaches to fraud analysis, based on advanced machine learning (deep neural networks) and mathematical modeling.
By establishing well-designed blocking and alerting policies in the Fraud Rules (for each fraud type), you can eliminate fraud with minimal possible losses, or even avoid fraud-related damages at all.
You can view this and any corresponding documentation page directly from the Control Centre by clicking the ? ABOUT button in the top-right corner of the page.
In the Fraud Rules module, you can set blocking and alerting actions for the fraud types that are available in your antifraud system version. You can find information about each fraud type (and its description) that the antifraud system is capable of detecting, including those that are not yet available in your version, in the corresponding module Fraud Types (described in the Information module group).
The algorithms of Proactive Monitoring can discover a single high-risk call attempt or a fraud attack preparation, which may include several fraudulent call attempts.
For the Behavioral Analysis method, when its AI detects a fraud attack, it determines the attack's parameters (including A-/B-number ranges of the attack).
Both methods deliver extremely high performance, with near-complete detection of fraud attacks, minimal detection delays, and an exceptionally low false positive rate.
The antifraud system uses over 200 fraud parameters to identify a fraud attack.
You can change the current settings for each fraud type by clicking the corresponding row (fraud type rule). This will take you to the Edit Rule page.
Here you will find the fraud type of the rule, and the corresponding configurable blocking and alerting actions.
The blocking action can be enabled or disabled using the toggle:
- If the blocking action is toggled off, no blocking action is applied to fraudulent call attempts of the specified fraud type;
- When the "Break and block calls" option is enabled, active fraudulent call attempts of the specified fraud type that match the corresponding A-/B-number ranges of the attack will be broken. A blockage is then set for those A-/B-number ranges for the specified duration (you can enter the blockage duraton manually or select it from the dropdown list). During this time, all further fraudulent call attempts that match the blockage criteria within the chosen window will be rejected.
The blockage window should be long enough to ensure that, after its expiration, fraudsters do not continue the activity.
Depending on your antifraud system version, the blocking actions may be unavailable. To upgrade your version, please contact your manager.
For some fraud types, the blocking action is not applied in the antifraud system, e.g. Flashcalls.
However, you can request a special flashcalls blocking feature to enable flashcalls blocking. In order to do that, please contact your manager.
Following that, you can enable or disable alerting to the specified Notification Channels:
You can adjust communication channels for alerts in the Fraud Alerts module of the Alerts & Reports module group.
After selecting the needed options, click the SAVE button to apply the changes, or the CANCEL button to cancel the changes and exit the Edit Rule page.