Behavioral Biometric (BehavioIQ)

When to use it

BehavioIQ (TrustVision's Behavioral SDK) collects device and user behavioral data for fraud detection and identity verification. It's available for Android, iOS, and Flutter, and is typically used to add a continuous, passive fraud signal — device fingerprinting plus touch/typing/sensor behavior — around events like login or a sensitive transaction, rather than a one-time onboarding check.

Key features

• Behavioral biometrics profiling
• End-to-end encrypted data collection, from device to server
• Touch, typing, and sensor data collection
• Configurable data collectors to match your app's needs
• Minimal integration effort

How it works

All platforms follow the same workflow:

1. Initialize the SDK in your app with your account credentials and the data-collection settings you want enabled.
2. Register the device — your backend registers the device once with TrustVision; this step is skipped on later sessions for a device that's already registered.
3. Start collecting — the SDK begins gathering device and behavioral signals (touch, typing, sensor motion) in the background as the user interacts with your app.
4. Track key moments — the SDK needs to know when the user moves between screens, and can optionally track individual text fields; touch is captured automatically.
5. Finalize the session once the behavior you want scored is complete — the SDK flushes any remaining data and hands you a session reference.
6. Get the score — your backend requests the score for that session and receives a risk assessment in response.
7. Clear on logout (optional) — reset the device identity when a user logs out, so the next session starts fresh.

See Behavioral API for the full endpoint reference behind each step, including authentication and error handling.

How scoring works

Behavioral scoring follows an enrollment/verification model: the first scored session for a user builds their behavioral baseline (a synthetic safe-default score, no real risk decision), and each later session is scored against that baseline. See the Behavioral API — Compute Behavior Score reference for the full behavior_score/risk_level mapping and the is_update override.