This article, written by the supplier feedzai, describes how machine learning evaluates the data in a payment transaction and can use that data to keep merchant fraud rates below the SCA exemption threshold. This in turn eliminates the need to implement two factor authentication of the shopper, which introduces friction that leads to cart abandonment.
The article describes all of the data elements that are available to merchants to help detect fraudsters. It does not, however, indicate that a broader data set generates more accurate results.
Using data from one merchant isn’t as effective as using data from 6,000 merchants, and data from 6,000 merchants isn’t as effective as data from every transaction on a given network. But as the dataset grows, the technology needed to maintain the performance of the machine learning tool becomes the primary challenge.
Updating the model takes longer, leaving a window of opportunity for criminals. In Mercator’s just completed comparison of eCommerce Fraud platforms, it became apparent that many of the machine learning platforms ingest data provided by others. Most platforms don’t have the ability to recognize and apply a risk metric to a new shopper as soon as they first touch the website, so these platforms ingest data from others, like Threatmetrix and InAuth, which deliver this capability.
Perhaps the networks, which have a stake in preventing fraud for issuers and merchants, should develop a method by which their machine learning results, derived by monitoring all payment transactions, can be shared with other fraud detection platforms.
In 2017, Google demonstrated Federated Learning, which enables mobile phones to collaboratively learn a shared prediction model while keeping all the training data on the device. Writ large, this would enable a worldwide fraud detection model that eliminates the need to share training data. Let’s dream big!
“Every transaction has hundreds of data points, called entities. Entities include time, date, location, device, card, cardless, sender, receiver, merchant, customer age — the possibilities are almost endless. When data is cleaned and connected, meaning it doesn’t live in siloed systems, the power of machine learning to provide actionable insights on that data is historically unprecedented.
Robust machine learning technology uses both rules and models and learns from both historical and real-time profiles of virtually every data point or entity in a transaction. The more data we feed the machine, the better it gets at learning fraud patterns. Over time, the machine learns to accurately score transactions in less than a second without the need for customer authentication.
Machine learning creates streamlined and flexible workflows
Of course, sometimes, authentication is inevitable. For example, if a customer who generally initiates a transaction in Brighton, suddenly initiates a transaction from Mumbai without a travel note on the account, authentication should be required. But if machine learning platforms have flexible data science environments that embed authentication steps seamlessly into the transaction workflow, the experience can be as customer-centric as possible.
Streamlined workflows must extend to the fraud analysts job
Flexible workflows aren’t just important to instant payments – they’re important to all payments. And they can’t just be a back-end experience in the data science environment. Fraud analysts need flexibility in their workflows too. They’re under pressure to make decisions quickly and accurately, which means they need a full view of the customer — not just the transaction.
Information provided at a transactional level doesn’t allow analysts to connect all the dots. In this scenario, analysts are left opening up several case managers in an attempt to piece together a complete and accurate fraud picture. It’s time-consuming and ultimately costly, not to mention the wear and tear on employee satisfaction. But some machine learning risk platforms can show both authentication and fraud decisions at the customer level, ensuring analysts have a 360-degree view of the customer.
Machine learning prevents instant payments from becoming instant losses
Instant payments can provide immediate customer satisfaction, but also instant fraud losses. Scoring transactions in real-time means institutions can increase the security around the payments going through their system before it’s too late.
Real-time transaction scoring requires a colossal amount of processing power because it can’t use batch processing, an efficient method when dealing with high volumes of data. That’s because the lag time between when a customer transacts and when a batch is processed makes this method incongruent with instant payments. Therefore, scoring transactions in real-time requires supercomputers with super processing powers. The costs associated with this make hosting systems on the cloud more practical than hosting at the FIs premises, often referred to as ‘on prem’. Of course, FIs need to consider other factors, including cybersecurity concerns before determining where they should host their machine learning platform.”
Overview by Tim Sloane, VP, Payments Innovation at Mercator Advisory Group
