Skip to main content

Beyond Rules and Attributes

Traditional fraud detection relies on rules and isolated entity analysis. It catches known patterns but misses the sophisticated schemes that exploit network relationships. Fraud is a network problem. A company might look clean in isolation, but its connections reveal the truth: shared addresses with known bad actors, ownership links to shell entities, suppliers with suspicious patterns.
Our Graph Foundation Model analyzes entities within their relationship context—capturing multi-hop patterns that traditional approaches can’t see.

How Avra Detects Fraud

1

Entity Intelligence

Registration anomalies, address inconsistencies, ownership patterns, activity mismatches
2

Network Analysis

Multi-hop relationships: counterparties, suppliers, ownership chains, shared infrastructure
3

Behavioral Patterns

Transaction velocity, payment patterns, seasonal variations—compared against similar legitimate entities
4

Temporal Dynamics

How is this entity evolving? Rapid changes in network position often signal fraud or distress

What We Catch

Signal TypeWhat Traditional Systems SeeWhat Avra Sees
Shell entitiesClean registration, no red flagsMulti-hop connections to known bad actors
Identity fraudValid documents, matching recordsOwnership network anomalies, address clustering
Bust-out schemesGood payment history buildingTrajectory toward known fraud patterns
Collusion ringsUnrelated legitimate entitiesHidden network connections, coordinated behavior

Use Cases

Onboarding Verification

Screen new counterparties before establishing relationships

Transaction Monitoring

Real-time risk signals for payment authorization

Portfolio Screening

Periodic review of existing relationships for emerging risks

Investigation Support

Deep network analysis when suspicious activity is detected

Powered by two foundations

Fraud detection is where Avra’s network intelligence is most direct. The Graph Foundation Model has learned what legitimate and suspicious network structures look like across the broader economy. Your Relational Foundation Model has learned the shape of your own transactions, customers, and accounts. The downstream fraud model is trained on top of both — and every training run feeds signal back into your RFM, making the next iteration sharper.

Customer Data Needed

DataPurpose
Fraud labelsHistorical confirmed fraud and legitimate cases to define your fraud definition
Transaction historyPayment patterns, amounts, and counterparty details
Application dataOnboarding information for registration-time scoring

Output Schema

FieldDescription
fraud_scoreProbability (0-1) that the entity or transaction is fraudulent
risk_factorsKey signals contributing to the score (network anomalies, velocity, entity attributes)
network_flagsSpecific multi-hop connections to known bad actors or suspicious clusters

Evaluation Metrics

  • PR-AUC — Primary metric, given the rarity of fraud events. Measures precision-recall trade-off across all thresholds.
  • ROC-AUC — Overall discrimination between fraud and legitimate activity.
  • False Positive Rate at fixed recall — Operational metric: how many legitimate entities are flagged at your desired catch rate.

Multi-resolution embeddings for fraud

Avra’s adaptive embeddings enable multi-resolution fraud architectures:
  • Transaction-level (64–128d) — lightweight embeddings for real-time transaction scoring where latency is critical. Fast enough to run on every payment.
  • Entity-level (512–1024d) — full-resolution embeddings for deep entity analysis. Rich context about the entity including network position, ownership patterns, and behavioral history.
This allows your fraud model to combine quick transaction signals with deep entity intelligence—smaller embeddings provide speed, larger embeddings provide context.