A composite risk score for every verification session.
ASTERIA KYC generates a composite risk score at the end of every verification session. Document result, face match confidence, liveness outcome, AML status, velocity indicators, and behavioural signals combine into a single 0–100 score with structured reason codes.
Composite risk score with explainable reason codes — calibrated to your policies, surfaced for every session.
Six input signals. One composite score.
Each signal contributes a weighted input to the final score. Signal weights are configurable per platform integration.
Document Result
Document authenticity outcome, integrity signal summary, and field extraction confidence contribute to the base document risk input.
Face Match Confidence
Biometric comparison similarity score and confidence band contribute to the facial identity risk input.
Liveness Outcome
Liveness result and confidence score, including detected attack type where applicable, contribute to the biometric presence risk input.
AML Status
AML screening result across sanctions, PEP, and watchlist datasets contributes a weighted compliance signal to the composite score.
Velocity Indicators
Submission frequency, device reuse signals, and session pattern anomalies assessed against behavioural velocity thresholds.
Behavioural Signals
Session metadata, interaction patterns, and submission timing anomalies contribute a behavioural risk modifier to the final score.
Structured risk output with full signal breakdown.
Every session returns a risk score, risk level classification, contributing factor breakdown, and structured reason codes for downstream decision logic.
| Field | Description | Values |
|---|---|---|
| risk_score | Composite risk score | 0–100 |
| risk_level | Risk classification band | low / medium / high |
| score_inputs | Per-signal contribution breakdown | object of signal weights |
| reason_codes | Structured reason array | array of strings |
| review_recommended | Human review recommendation | true / false |
| score_version | Scoring model version | string |
{
"risk_score": 46,
"risk_level": "medium",
"score_inputs": {...},
"review_recommended": true,
"score_version": "v2.1.0",
"reason_codes": ["VELOCITY_ELEVATED"]
}Three risk bands. Configurable thresholds.
Risk level thresholds are configurable per integration. Default bands provide a starting baseline for onboarding decision logic.
Low Risk
All input signals returned within acceptable ranges. Risk score below the medium threshold. Standard onboarding flow proceeds without review escalation.
Medium Risk
One or more signals returned elevated readings. Risk score in the medium band. Case may be routed to human review based on platform configuration.
High Risk
Multiple signals returned above threshold or one critical signal failed. Risk score in the high band. Session escalated for review or rejected.
Ready to turn identity verification into a controlled compliance workflow?
Use ASTERIA KYC to verify users, screen risk, and preserve evidence from one connected platform.