Device data provides critical information about the hardware and software environment a user is using, helping to identify suspicious activities and potential fraud.

​
Device Data

Verisoul collects various device signals to create a comprehensive profile of the user’s environment. This data helps in identifying legitimate users and detecting potentially fraudulent activities.

​
Example Device Data

{
  "browser": {
    "type": "Chrome",
    "version": "132.0.0.0",
    "language": "en",
    "user_agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36",
    "timezone": "America/Chicago"
  },
  "device": {
    "category": "desktop",
    "type": "Mac",
    "os": "macOS 10.15.7",
    "cpu_cores": 16,
    "memory": 8,
    "gpu": "ANGLE (Apple, ANGLE Metal Renderer: Apple M4 Max, Unspecified Version)"
  }
}

​
Data Collection Limitations

It’s important to note that the device information collected is not guaranteed to be complete or accurate due to several limitations:
  • Privacy-preserving browsers: Browsers like Brave, Firefox, and Safari may intentionally mask or limit access to certain device characteristics.
  • Memory reporting: Some browsers restrict access to accurate memory information.
  • User agent spoofing: Users can modify their user agent strings to appear as different browsers or devices.
  • Hardware access restrictions: Browser security measures may limit access to detailed hardware information.
  • OS-level restrictions: Operating systems may prevent access to certain device characteristics.
These limitations are part of the evolving privacy landscape in web browsers and should be considered when analyzing device data.

​
Device Risk Score

The device risk score is a machine learning model output that evaluates how likely a device is to be associated with fraudulent activity.

​
Example Device Risk Score

{
  "device_risk": 0.0295
}

​
Understanding Device Risk

The device risk score is triggered when characteristics about the device deviate from patterns seen in normal devices. Higher scores indicate greater likelihood of suspicious activity. Factors that can influence the device risk score include:
  • Browser automation: Signs that the browser is being controlled programmatically
  • Bot activity: Patterns consistent with automated bot behavior
  • Emulators: Evidence that the device is an emulated environment rather than physical hardware
  • Virtual machines: Indicators that the session is running in a virtualized environment
  • Inconsistent device signals: Contradictory or impossible combinations of device characteristics
  • Tampering: Signs that device fingerprinting is being actively circumvented