Digital Identity Graph helps detect fraud, cyberattacks

ThreatMetrix announced at RSA Conference the newest innovation in the fight against global cybercrime – the Digital Identity Graph, a framework for anonymized global digital identities on the internet.

The Digital Identity Graph maps the complex and ever-changing associations among people and their communication devices, account credentials, telephone numbers, physical addresses and the businesses with which they interact. Using the underlying graph framework, ThreatMetrix can define digital persona relationships on a global scale, enabling businesses to identify fraud and cyberattacks with real-time speed and precision without compromising privacy.

The Digital Identity Graph framework is a data science and engineering achievement with multiple far-reaching applications. ThreatMetrix anticipates that these concepts will be adopted broadly and yield enduring value for the security and fraud prevention industry as a whole.

Today, the Digital Identity Graph is the underlying fabric that powers global shared intelligence through the ThreatMetrix Digital Identity Network. ThreatMetrix applies machine learning, behavioral analytics and other advanced techniques on top of the graph across 2 billion monthly transactions serving 4,000 customers globally. The graph is strengthened with every transaction through the ThreatMetrix network, mapping the relationships of digital identities across the globe.

“The Digital Identity Graph allows us to leapfrog established authentication procedures that are widely known in the analog world,” said Andreas Baumhof, chief technology officer. “For example, the traditional credit bureau model of authentication based on knowing a name, physical address or social security number is rapidly becoming outdated and irrelevant. This static information is already known to fraudsters due to numerous data breaches. The power of the graph is knowing when your information is used out of context – without even needing to know who you are. In milliseconds, we can differentiate between a cyber attack and a trusted user without compromising privacy.”