Voice of the Industry

How a top 5 Bank uses machine learning to detect fictious identities

Friday 18 October 2019 00:13 CET | Author Melisande Mual | Voice of the industry

Detecting the use of stolen and synthetic identity credentials early on the customer journey, requires the use of adaptive decisioning. Simility shares how their industry recognized, best-in-class machine learning platform helped a top bank combat new account fraud

From opening a bank account to applying for a mortgage, millions of people submit applications to thousands of banks around the world every day. Processing these applications and ensuring that the people applying are legitimate customers is a daunting challenge – a challenge that is compounded by the rapid digitalization of bank services. Mobile banking has all but replaced in-branch visits, and applying for a loan now only takes a few clicks.

In this era of unprecedented digital access, banks must rapidly innovate to meet the current and future demands of their customers. As such, banks have prioritized rolling out slick online experiences that outshine their competitors. However, building a robust and competitive digital offering is an intensive and taxing process, and has left serious gaps in the aging infrastructure of banks. Unsurprisingly, these gaps are targeted by fraudsters, and with readily available tools from the dark web, easily exploited.

At the same time, countless data breaches have exposed the personally identifiable information (PII) of millions of individuals. Using stolen or synthetic identity credentials, fraudsters can pass a majority of checks against third-party validation services to apply for credit, loans, benefits, and more before ultimately defaulting on payments – leaving credit and lending companies with massive financial losses.

As fraudsters continue their attempts to monetize stolen and synthetic identity credentials while safely masked behind the Internet, businesses need to prevent fraudulent account applications from entering their ecosystem and causing substantial damage later on in the customer journey.

vspace=2

Unmasking Fraud with Advanced Machine Learning

One of the top five largest banks in the United States was having difficulty differentiating between legitimate and fraudulent new account applications. Because fraudsters were able to slip into the bank’s ecosystem and evade detection, they were able to accrue and default on massive amounts of debt. The bank needed a robust fraud and risk management platform that could detect suspicious activity early on in the customer journey as well as uncover emerging fraud patterns, without impacting good customers.

Upon implementing Simility, the bank discovered fraudulent activity occurring on its online and mobile channels. Fraudsters were using stolen and synthetic identity credentials to apply for new bank accounts, which they then used to funnel large amounts of money generated from illegal activities. Through Similitys robust link analysis and machine learning capabilities, the bank was able to detect and block anomalous behavior that connected nearly 60 fraudulent bank accounts worth millions of dollars.

Complex Fraud Requires Adaptive Decisioning

Similitys Adaptive Decisioning Platform combines various static and dynamic data sources, with advanced machine learning to help unmask even the most sophisticated attacks. Simility’s powerful device fingerprinting technology analyzes hundreds of mobile and desktop device characteristics, then uses cluster analysis and fuzzy matching to help alert the bank to suspicious activity before cybercriminals even have a chance to complete an account application.

vspace=2

The platform’s practical user interface enables the bank’s analysts or other non-engineers to quickly and easily add, configure, modify, and optimize rules and machine learning models. Advanced machine learning and big data analytics help the bank continuously enhance the detection of anomalous patterns based on changing events. Robust link analysis and visualization capabilities further help the bank investigate and connect data points to reveal patterns and relationships indicative of complex fraud schemes.

Simility continues to help support the bank’s commitment to providing financial products and services that make banking safe, simple, and convenient. With Simility, the bank is able to balance competing priorities, including fraud detection, customer experience, and regulatory requirements.

For more information on how to use adaptive decisioning to prevent new account fraud, download the solution brief here.

About Simility

vspace=2Simility offers real-time risk and fraud decisioning solutions to protect global businesses. Simility’s offerings are underpinned by the Adaptive Decisioning Platform, which is built with a data-first approach to deliver continuous risk assurance. By combining artificial intelligence and big-data analytics, Simility helps businesses seamlessly orchestrate decisions to reduce friction, improve trust, and solve complex fraud problems.


Free Headlines in your E-mail

Every day we send out a free e-mail with the most important headlines of the last 24 hours.

Subscribe now

Keywords: fraud, fraud detection, NAO, new account opening, new account fraud, banking, financial institutions, risk assessment, Simility, machine learning, Adaptive Decisioning Platform
Categories:
Countries: World






Industry Events