CO-OP Financial Services has brought in a new data-driven platform to help US credit unions detect and tackle fraud more quickly.
The organisation, which operates an interbank network connecting the ATMs of credit unions in the US and Canada, says its COOPER platform uses machine learning to keep up with patterns of online crime.
President and chief executive Todd Clark said the multi-million dollar project, CO-OP’s largest technology initiative, would bring “greater security across our products and services, while providing the most seamless experience to the member.”
He added: “The machine learning power of COOPER will allow us to constantly improve upon the fight against fraud by enabling the understanding of huge amounts of data and detecting complex patterns rapidly.”
Positioned behind the scenes as a data-driven “brain”, COOPER will be used exclusively for the company’s client credit unions. Data will be harnessed to incorporate machine learning and eventually AI to better mitigate fraud and enable greater member engagement.
Beginning this summer, CO-OP will roll out COOPER Fraud Analyzer, an account-based risk management solution for CO-OP Shared Branch clients that will provide added protection to identify and prevent potentially fraudulent transactions.
The company will then launch COOPER Fraud Score, using machine learning to create a risk-scoring model that determines the level of suspicion for a transaction. This capability will be available on account-based products later this year, and for card-based products in 2019.
CO-OP says this means “the power machine learning will run across the entire CO-OP ecosystem, further integrating the user experience for clients and members”.
Mr Clark added: “Understanding members and predicting patterns also enables more targeted communications and offer. COOPER is being introduced as an enhancement to our fraud mitigation solutions, but it will be additionally leveraged later for greater member engagement and more precise targeting to anticipate member needs.”
Machine learning can reduce manual review of queues, easily adapt to new business lines, reduce false positives and increase precision. CO-OP’S advanced fraud products, including COOPER, card controls and credit line risk tools, are designed to adhere to customised fraud tolerance levels by individual clients.
