Multi-Model Risk Assessment Engine
Ensemble weighting fusing supervised fraud, behavioral anomalies, and AML graph typologies.
Ensemble Scoring & Tier Mapping
Deterministic formula mapping raw multi-model inputs to 0–100 risk scores.
Ensemble Weighted Logic Configured in Department 6
Contract VerifiedThe risk engine schema is integrated into backend contracts. The formula (0.45*fraud + 0.25*anomaly + 0.30*aml) and hard override rules will produce calibrated risk tiers.
Module Architecture Contract:
Ensemble Formula:0.45·Fraud + 0.25·Anomaly + 0.30·AML
Override Rule:Detected AML cycle forces CRITICAL tier
Tiers Defined:Low (<30), Medium (30-59), High (60-84), Critical (85+)
Endpoint Status:GET /api/v1/risk/status (HTTP 200 OK)