Generating Unique and Powerful Insight
As individuals use their identities to apply for goods and services – credit, loans, wireless phones, mortgages, etc. – certain patterns emerge. ID Analytics applies advanced analytics to understand these complex information relationships and deliver a unique perspective on the opportunity and risk associated with U.S. consumers.
By applying advanced analytics to data within the ID Network®, ID Analytics quantitatively evaluates millions of desirable and suspicious behaviors and relationships, in real time, to understand identity risk. These analytics generate unique and powerful insight into the authenticity of an identity, the creditworthiness of a consumer, or the likelihood an application represents an attempt to commit fraud.
We have developed proprietary capabilities that help our solutions address a number of challenges around identity and risk. Examples include:
- Personal Topology: We can better analyze a person by studying connections. Each identity element provided in a consumer interaction, like an address, has its own history of connections to other identities and events. By looking across these connections, ID Analytics is able to apply a broad lens to risk assessment, accurately assessing the opportunity and risk represented by a wide range of consumer behaviors. Personal topology is a core technology embedded across all of our solutions.
- Identity Resolution: With this capability, we are able to pinpoint a person even when only fragments of their identity are visible. We use a combination of complex data synthesis, architecture and specialized fuzzy matching algorithms on difficult fields such as name and address.
- Machine Learning: The algorithms behind our products come from the state-of-the-art machine learning community, and include support vector machines, neural nets, boosted trees, random forests and both supervised and unsupervised clustering algorithms. By incorporating the best aspects of these and other techniques and customizing these algorithms to fit our data and business problems, our experienced modelers continue to deliver superior performance.
- Validating Never-Before-Seen SSNs: Now that Social Security numbers are randomized and no longer follow a published formula, you would expect that there would be no way to evaluate the legitimacy of never-before seen SSNs falling outside of previously known ranges. By constructing a complex series of algorithms, ID Analytics is able to provide insight into these increasingly common SSNs, a capability utilized in the latest version of ID Score®.