New pattern analysis techniques to defend against fraud

Sophisticated fraudsters scale systems to increase their ROI. But it’s also a weakness that you can exploit to shut down fraud rings and keep attacks from scaling.

In this discussion, fraud experts Nisreen Hussain, Irfan Faizullabhoy, and Ashley Fang show off how pattern and link analysis stop AI-powered fraud, account takeovers, and large fraud rings.

WHAT WE COVER
→ The three layers of fraud detection: image checks, environment signals, and population-level link analysis
→ How Persona's Graph connects accounts and improves automation
→ What it takes to build real-time link analysis in-house: graph database latency, image vectorization, and text normalization (address tumbling, email tumbling)
→ How Handshake AI uses graph analytics to detect coordinated fraud rings and enforce against entire clusters at once
→ Live demo: discovering 25 linked accounts from a single suspicious verification, then building a rule to flag future attempts automatically
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
SPEAKERS
Ashley Fang — Fraud product analyst, Persona
Irfan Faizullabhoy — Product lead (link analysis), Persona
Nisreen Hussain — Head of policy & content moderation, Handshake

#FraudDetection #LinkAnalysis #FraudRings #GraphAnalytics
#IdentityVerification #FraudPrevention #PatternAnalysis #FraudIntelligence