The Fraud Network Detection tool is a complex fincrime typology agnostic solution for clustering applicants based on similarities in their behavior and patterns of using technical capabilities. It significantly enhances control and helps you investigate different techniques across fincrime types.

The whole process is built and designed including machine learning technology for ensuring constant improvements and faster detecting fraudulent activities at both KYC and KYT stages of verification.

When launching the tool, you will be able to access the following benefits:

• Prevent multi-accounting. Through network analysis, we unveil connections between accounts using similar templates or originating from the same IP addresses or devices.
• Boost deepfakes recognition. Uncover deepfakes looking for additional clues and use information on the found fraud networks to detect the whole clusters of deepfakes.
• Combat chargebacks with enhanced mule detection. Development in detecting fraudulent networks allows us to actively participate in the ongoing arms race against deepfakes and as well combat against the mules, which are a popular form of fraud where another person's identity is purchased instead of being generated or stolen.
• Monitor traffic. By assessing completion speeds, geographical patterns, and other relevant parameters, we distinguish genuine user engagement from artificially motivated traffic, enhancing the overall reliability of your platform's user base.
• Detect bot farms. Via analysis of the device fingerprints, completion speeds, and other behavioral nuances, with Sumsub's Fraud Network Detection, we efficiently detect and neutralize the threat posed by bot farms, safeguarding your platform's integrity.

How it works

Sumsub's Fraud Network Detection shows connections between the applicants/clusters of applicants based on various parameters. Links between the applicants help you indicate suspicious activity and proactively investigate and intervene in criminal networks without relying on money movement as the sole trigger. According to gathered information, your applicants can be clustered as follows:

• By device, IP address, cookies, Liveness device (Camera) and the address specified in PoA documents.
• By similar documents and selfies.
• By action time (e.g. many applicants related to the same device have been created in a short amount of time).

We calculate networks every day for clients with at least 500 applicants for the last 30 days with added documents or selfies through WebSDK or MobileSDK.

Get started

Almost nothing is needed to start detecting potential fraudulent patterns amongst your applicants:

1. Contact your Customer Success Manager to implement Fraud Networks Detection in your verification process.
2. Review the results on the criminal connections detected when verifying applicants.

📘

Use WebSDK

The more data is shared with Sumsub and the more our services are used, the more illegal clusters can be detected, so WebSDK integration is preferable for such cases.

Review results

You can find the results on the Fraud Network Detection findings in the following ways:

• Use the Sumsub API:
  • To get the list of all fraud networks, use this method.
  • To get the list of all fraud networks by applicantId, use this method.
  • To get the fraud network by networkId, use this method.
• In the Dashboard, go to the Applicants menu and select the Applicant Networks tab. Here, you can see the list of fraud networks associated with your applicants.
• In an applicant profile, scroll down to the Applicant Networks block and find information on the fraud networks related to the applicant.

The following table describes network properties available in the Dashboard:

Results can be downloaded as a CSV report as well and will contain information on all applicants included in the network. The following data will be available in the report:

Network views

When checking the network data, you can switch between the table and chart views: