Introducing Incremental Learning: An Industry-First Boost for Fraud Prevention
In our previous blog on machine learning, we sought to clarify its role in fraud prevention for merchants. To summarize, it can be an extremely effective way to identify patterns of fraud in a manner and at a speed that humans cannot. It is a critical tool in the fight against fraud, especially when used as part of a multi-layered fraud solution.
There are, however, challenges around the development and training of machine learning models that merchants need to understand:
- Fraud represents a small fraction of all daily transactions
- The distribution or pattern of fraudulent and good transactions evolves over time, due to seasonality, changes in customer behavior and new fraud attack strategies
- Without timely checks, the true nature of most transactions is usually only revealed a few days – or even weeks – after they actually take place
What this means is that the training of machine learning models requires a very large amount of historical transaction data. And, as good and bad behaviors change, the performance of most machine learning models will degrade. They will need to be retrained on large amounts of data.
Enter incremental learning...
We recently announced the launch of incremental learning capabilities to address this challenge. This exciting new approach to modeling is designed to ensure that machine learning models perform to the highest quality for longer than current traditional models.
Incremental learning models are able to “think for themselves” and make small adjustments on an ongoing basis, ensuring they remain relevant even as fraudsters and genuine consumers change their behaviors. This represents a significant advancement over traditional machine learning models that need to be retrained as fraud patterns change.
Learning the new tunes
To understand how incremental learning works, think of it like playing the piano. As we get better at playing the piano, we learn new tunes. We build our repertoire over time – and we don’t need to re-learn all the old tunes when we learn a new one.
This is how an incremental learning model works – and this is the point: a traditional machine learning model would have to re-learn all the tunes it previously knew to be able to learn the new one.
Incremental learning algorithms allow a model to adapt to new behaviors without needing to re-learn everything it already knows. New data is fed to the algorithm every 24 hours and it spots new behaviors as they happen. If the change in behavior is enough to cause a degradation in performance, the model will adapt to what it has recently learned. This ensures that model performance quality lasts for longer periods of time, without degradation. In tests, our incremental learning models maintained their performance over 13 months, while others started to degrade after just three months.
What this means for merchants
Incremental learning reduces the need to retrain over large data sets, and the need to continuously re-deploy to production, because the changing data is incorporated on an ongoing basis. That reduces the demand on scarce merchant resources.
Most importantly though, incremental learning performs better than traditional methods. By using the most up-to-date fraud intelligence and customer behavior data, and continually adjusting in response to that data, models can protect merchants more effectively, over longer periods of time.
Given the continuous evolution of customer preferences, shopping trends and fraudster behavior, assessing transactions against the latest data has never been more important. Fraudsters will continue to adapt their tactics and techniques in an effort to evade detection – incremental learning will be instrumental in keeping up with fraudsters and stopping them in their tracks.
Watch our on-demand webinar: Find a New Ally in the Fight Against Fraud with Machine Learning
Related Blog Posts
An Executive’s Guide to the Top Five Fraud KPIs
Key performance indicators (KPIs) are a commonly used term across many industries. For KPIs to be a valuable way to measure success, they need to be properly defined, capable of being adjusted and connected to one another. When it comes to use of KPIs in fraud management, this is particularly important. With this in mind, let’s take a closer look at the different fraud KPIs, along with how they should ideally interact.
Real-Time Payments: A Mix of Simple, Smart and Secure Will Be Crucial for Further Adoption
In our global payments report, we predicted that 2020 would be Prime Time for Real-Time. And that was before the COVID-19 pandemic accelerated the adoption of real-time and digital payments around the world, which quickly became a necessity in keeping citizens safe and businesses afloat. During the recent ACI Edge Virtual event for banks and intermediaries, we discussed the challenges of real-time adoption with some of the industry’s leading experts. I caught up after the event with Lisa Neary, head of payments at Co-op Bank UK, Ian Gausden, head of real-time payments at Mastercard, Vincent Brennan, head of group payments and business continuity management at Bank of Ireland and Enrico Albertelli, head of payments and digital banking solutions at Nexi.
Merchants Don’t Need a Payments Gateway, They Need a Payments Hub
The term “payments gateway” over-simplifies what it takes to process a payment. The terminology comes from the early days of online payment processing, but has now entered the vernacular for all aspects of merchant payments, including online, mobile and in-store.
How to Recession-Proof Your Retail Business in 2021
2020 has been a challenging year for retailers, many of which who were — understandably — unprepared to deal with the pandemic. With millions of people in the U.K. in lockdown, consumers and businesses have had to change the way they work and live, and the way they shop and pay.
What Will the World of Post-Pandemic Payments Look Like? [Dave Birch Q&A]
Dave Birch is a leading global authority on payments and digital identity, who is no stranger to predicting what the future of financial services has in store. After delivering the keynote presentation at our recent ACI Edge Virtual: Banks & Intermediaries, we gathered some insights from Dave on what the world of payments could look like, post pandemic.
12 Ways Americans Changed How They Pay in 2020
ACI experts recently shared their predictions and industry expectations for billing and payments in 2021 and beyond. As we close out 2020, let us look back at just how much billing and payments have changed this year:
Comment DXC Technology et ACI Worldwide adressent-ils les enjeux règlementaires en Europe?
En Juillet dernier, Le conseil de gouvernance de l’ECB a pris une décision importante concernant le futur des paiements en Europe. En demandant aux PSP participants au système TARGET2 de se connecter à TIPS avant Novembre 2021. Dans le même temps les ACH domestiques devront effectuer le transfert de leurs comptes techniques vers TIPS.
Adding Fuel to the Fraud Fire
Payments fraud is a widespread and difficult issue in the fuel segment. Fuel dispensers are a common target for certain types of fraud – and fuel merchants often lack visibility into the level and impact of that fraud.
EBAday 2020: Reaping the Full Potential of Real-Time Payments
Real-time payments are growing rapidly worldwide. In fact, according to ACI’s Prime Time for Real-Time report, more than half a trillion real-time payments transactions will be processed over the next five years. ACI’s Dean Wallace, director solution leadership, Real-Time and Digital, recently took part in an EBAday panel to discuss what businesses need from banks to reap the full potential of real-time payments, how fintechs can assist and how collaboration can be advanced on a global scale. I caught up with Dean to discuss the latest insights.
The Role of Regulation, Collaboration and Competition in Pushing Forward Payments Modernization
While geographic regions have taken their own approaches to payments modernization, they all face some common challenges – a principal one being adoption of real-time payments, which is central to successful payments modernization projects. I spoke to some of the top minds tackling real-time payments adoption in key markets, including the U.S., Canada and the U.K.