System increase fraud detection rates, reduces false positives
New fraud-busting software will improve detection rates by going beyond basic transaction analysis to scrutinise seemingly innocuous aspects such as shared address, telephone numbers and account ownership, according to business analytics software supplier SAS.
The vendor has produced its SAS Fraud Framework to prevent and detect various types of fraudulent activities in banking, insurance and government organisations.
The system includes the SAS Social Network Analysis software which can be trained to help find previously unknown relationships that by themselves do not seem suspicious, but in concert are clearly fraudulent.
The software should allow investigators see connections more clearly so they can uncover previously unknown relationships and conduct more effective and efficient investigations, the company said.
In addition to detection and risk scoring, investigation teams can review visualisations of relationships that include individuals flagged by existing rules, anomaly detection or predictive modelling.
Useful for money laundering detection operations, it is claimed that the system should increase fraud detection rates and reduce false positive of systems used to monitor customer behaviour across multiple accounts and systems.
At the core of the framework is a profiling engine that scores individuals, accounts, products and networks based on rules, fraud scores and links to known fraudsters.
John Brocklebank, VP for SAS Solutions OnDemand said the software has many uses beyond fraud detection and prevention. “Businesses such as telecommunications companies and banks can use network analysis to better understand customer behaviour and target relevant offerings to new and existing customers.”
The SAS Fraud Framework becomes available this month.