ACI Worldwide Launches Advanced Machine Learning Capability to Dramatically Reduce Payment Fraud for Merchants
ACI Worldwide (NASDAQ: ACIW), a leading global provider of real-time electronic payment and banking solutions, today announced the launch of its new Incremental Learning technology, an innovative industry-first approach to machine learning that will enhance fraud protection for merchants and financial institutions and their customers.
ACI has filed a patent application for its "Incremental Learning" technology, which is being implemented in machine learning models within ACI's fraud prevention solutions. These include Proactive Risk Manager, which delivers enterprise fraud management capabilities for financial institutions and intermediaries, and ACI ReD Shield, which underpins ACI's secure eCommerce and merchant fraud management capabilities.
ACI's Incremental Learning technology represents a significant advancement over current machine learning models that need to be retrained as fraud patterns change. Incremental Learning models are able to "think for themselves" and make small adjustments on an ongoing basis to ensure they remain relevant, even as fraudsters and genuine consumers change their behaviors. Tests carried out on data from three major retail customers over 13 months revealed that, while traditionally trained models began to degrade after three months, ACI's incremental models maintained their performance over the full period of the test.
"Traditional machine learning models in many cases are not sufficient to stop fraudsters in their tracks. As fraudsters become more sophisticated, we need to continuously advance our models to beat them at their own game," said Jimmy Hennessy, director of Data Science at ACI Worldwide. "Our global data science team has created a game-changing piece of machine learning technology that can be seamlessly integrated and future-proofs the precision and operational efficiency for over 5,000 institutions protected by our solutions today."
ACI has more than 20 years of experience in designing and implementing machine learning models, which have long been a fundamental element within Proactive Risk Manager and ACI ReD Shield. ACI's machine learning models will quickly and efficiently analyze all available features and data points, which will then be turned into intelligence that can build customer profiles, spot fraud signals, and combat emerging fraud threats. Any anomalies are flagged in real time and immediately actioned.
"We are the first vendor globally to roll out the new Incremental Learning technology across the merchant, payments, and financial services sectors," commented Fabian Gloerfeld, head of Payments Intelligence, ACI Worldwide. "The new capability is a realization of ACI's multi-year investments and will further enhance our sophisticated fraud monitoring and prevention solutions to help customers to dramatically reduce payments fraud."
Find out more about ACI's Incremental Learning -- A Smarter Way to Fight Fraud at: https://www.aciworldwide.com/capabilities/machine-learning/incremental-learning
Host a Webinar with the MRC
Publish Your Document with the MRC
There are no related Events
Manual Review remains a cornerstone of modern fraud prevention; offering enhanced precision in fraud screening, fairness in dispute management and labels for machine learning. However, as a business grows, so to must a Manual Review team. This introduces operational challenges to manage headcount, quality and return on investment.
In this session, we will cover Booking’s story of launching and rapidly scaling Manual Review Operations as a Service in Fraud Prevention. Hear how we empower a small team of 4 to manage Manual Fraud Operations on a global scale across Booking’s suite of products and associated risks.
As the need to reduce the cost of manual fraud prevention increases, we will show the benefits of leveraging multiple fraud prevention solutions such as outsourcing, automation and ML, demonstrating how the function interacts with various ML models to optimise workload and achieve a balanced defense against fraud.
Conventional fraud detection has historically been centered around linear relationships between user indicators. Machine learning techniques in this domain also focus on learning and adapting to patterns arising from single depth connections of datapoints. The weak point of this setup is that we miss out on transitive relationships, which becomes an increasingly interesting factor when fraudsters reuse few overlapping indicators in new fraud attempts.
At Booking.com, we solved this problem by using graph technology to power our payment fraud controls. In this presentation we will show how we developed an in-house technological ecosystem that stores payments transactions in graph format and leveraged state of the art algorithms to compute innovative graph features that enriches our existing fraud controls (ML models and static rules). We will have a look at the different parts of the system: a latency sensitive real time feature computing service, a visualization tool for analysis of networks and a historical feature reconstruction mechanism. Finally, we will have a look at the impact we saw on our fraud controls after incorporating the new features.
We would conclude by explaining the plans to scale this technology in other fraud domains like marketing & rewards, account takeover etc. and exploring Network Representation Learning (a type of deep learning on graph data) to tackle them.
Business leaders at global marketplace Wish knew the secret to the enterprise's continued growth was increased international expansion.
As one of the world’s largest e-commerce marketplaces, Wish now operates in more than 60 countries, processing approximately 900,000 transactions a day, across about 250,000 merchants, and has more than 20 million monthly active users. While that scale provides a wealth of transaction intelligence, Wish also needed to put a very local lens on the data it was processing.
Wish found its answer in machine learning — in particular a machine learning fraud protection model that scales rapidly while leveraging the fraud and risk expertise of the internal team at Wish. Smart machines surfaced the localized data the Wish team needed to react to the diversity of geo-specific fraud attacks and abuse schemes.
In this session, Signifyd senior vice president of operations and corporate development J. Bennett and Wish Director of Risk Operations Tara Mitchell dive into the considerations around deploying a machine learning solution and explore the steps Wish took to marry the strengths of humans and machines to reduce their exposure to fraud and increase order approval and conversion rates.
Machine learning continues to be an important tool for fraud mitigation specialists, but fraudsters are learning how to adapt.
The current landscape is now a battle against both automated and human-driven attacks. This presentation from Adyen and Microsoft explores the latest techniques in AI and ML to ensure the best customer experience and provide additional context for fraud decisions.
There are no related Surveys
False positives down, revenue up! Learn from an Experian fraud expert how machine learning strengthens fraud prevention, reduces false positives, and leads to new revenues.
Fraud prevention is one of the most exciting areas in commerce. However, it is also challenging to core business functions. This webinar will present how modern fraud prevention becomes more efficient through machine learning models and how this results in new revenue potential.
By participating in this webinar, the learner should be able to understand:
- State-of-the-art fraud prevention methods and current fraud figures
- How machine learning supports fraud prevention
- Why companies should rely on smart fraud prevention with machine learning models
We have met with many cases where analysts not only doubt themselves in their decision but also discouraged to take information into account that may not be considered as useful – not to mention the rise of Machine Learning in fraud prevention. Is our work that we do manually truly outdated?
Through examples we will re-explore the tools that we have at our expense and discuss how we can effectively use them in relation, from articulating sentences, to defining use cases and relying on AI.
- To find or regain the interest in the beauty of fighting fraud – making decisions that would be the key to stop the malicious activity yet keeping the mind open to changes.
- Utilising every aspect of the tool that are available and considering ways that we may have disregarded before and still could be useful, ultimately a boost for the year 2022 and to remain on the top of defense against harmful users.
In this thought-provoking session, we will share best practices on combining AI and ML with some of the more traditional approaches to mitigating online fraud and highlight why extensive cross-industry collaboration can help organizations strike the optimal balance between security and usability.
Types of fraud prevention systems which are payment-agnostic are addressed next, followed by a brief analysis of machine learning models which use no payment data. The broadcast concludes with a detailed Q&A session.
At SEON Technologies we have released new information on the collection countries that are most and least at risk of cyberattacks. We have also taken a close look at the most common types of cybercrime occurring in the US.
Dubbed the Global Cybercrime Report, the report explains how several countries are the safest in the world from fraud and other cybercrime. and why others are not. Our methodology for this research was based on how companies and public infrastructure are all being fairly well protected through both legislation and technology at their disposal.