Detecting and stopping fraud in today’s fast moving payments landscape is a job which requires accurate, real-time screening and decisioning capabilities that can adapt to changing trends.
To meet this demand for speed and accuracy, fraud management solutions must be underpinned by the ability to quickly and efficiently process and analyze vast amounts of data — turning it into intelligence that can build customer profiles, spot fraud signals and combat emerging fraud threats.
Enterprise-class adaptive machine learning is a core component of ACI’s UP Payments Risk Management solution — a product of over 20 calendar years of machine learning algorithm, analytics process and software design and development experience.
Working closely with our extensive network of banking, intermediary and merchant customers, we have significantly developed these capabilities over the last two decades, ensuring that our suite of models, products, strategies and processes has evolved to address changing behaviors, market trends and developments in technology.
Optimize Your Fraud Strategy Through Machine Learning
When applied to the task of fraud detection, machine learning has the ability to learn from past customer information and confirmed fraud intelligence to identify current fraud patterns and highlight emerging trends that are otherwise too complex to be noticed by humans.
At ACI, our machine learning models are adaptive, ensuring that the algorithms we use continuously incorporate feedback and adapt to change — supporting accuracy and fast action against fraud without needing to repeat the training process.
ACI's Fraud Prevention Capabilities Include Several Different Approaches to Deploying Machine Learning Models
ACI can offer segment models to individual merchants or banks, or a small segment of businesses, to enable them to achieve higher fraud detection rates at lower false positive rates.
Segment models are trained on only the specific customer’s transaction data, it also uses shared fraud intelligence from a cross-section of other relevant organizations to ensure a wider context for the model training process.
This enables ACI to optimize model performance on a single customer’s data, while ensuring that customer still benefits from ACI’s global view of payment and fraud trends.
Trained using transaction data and confirmed fraud intelligence from multiple merchants within a single industry, our sector models also benefit from ACI’s global view of card-not-present customer and fraud transaction patterns.
This approach is vital to achieving higher predictive performance, since it offers a window to fraudsters’ activity as they operate across merchants, IP addresses, email addresses, devices, channels and geographies.
Sectors such as telcos, gaming and other specialized merchant categories particularly benefit from this type of approach because the models are more closely configured to address the nuances of that particular sector.
Developed on the full dataset of a financial institution, our custom models benefit from a global view of the customer’s genuine and fraud transacting patterns, which is important to achieving higher predictive performance since fraudsters operate across all bank and intermediary products and channels.
We optimize our custom models to detect specific segments of fraud, segments which achieve the overall business and operational requirements for financial institutions.
Extend Your Strategic Fraud Management Bench Strength
Collectively this team has over 80 staff-years of experience applying the use of machine learning algorithms in payments, banking and retail fraud detection.
The Data Sciences team is supported by ACI’s global team of risk analysts and fraud consultants. Both departments work closely together, and with our customers and partners across the payments ecosystem, to constantly monitor the performance of our machine learning models and review shared intelligence to stay one step ahead of the fraudster.
It is this approach that puts ACI ahead of the market for delivering highly effective fraud strategies that protect our customer’s businesses and enable best-in-class consumer payment experiences.