Innovation in financial services: Credit card fraud detection

Banking leaders need to modernize the customer experience, grow their customer base, and reduce operational costs. To achieve these goals, banking institutions must continue to integrate data from various sources and translate that data into actionable insights while meeting complex regulatory requirements. IT investments in cloud infrastructure must accelerate the data journey to enable business innovation, personalize the customer experience, and increase operational efficiency.

In this article, we focus on credit card fraud detection. With this necessary business function, banks can target technology investments to protect their resources and their reputation, while giving customers a better experience.

Credit fraud and how Oracle can help

Credit card fraud is on the rise, costing the financial industry $28.6 billion in 2020, according to the Federal Trade Commission. Fraud results in loss of resources and puts a bank’s reputation at risk.

Credit card fraud is the most common and costly attack by fraudsters. As banks expand their digital footprint the potential attack surface also expands, resulting in more vulnerabilities. Banks have responded by developing advanced fraud detection techniques that deliver increased precision, reduced recalls from false positives, improvements to the customer experience, reduced administration during claim filing, and faster speed-to-resolution.

Fraud patterns change over time, and traditional deep learning methods don’t always detect rare events in very large data sources. Here, Oracle’s analytics and artificial intelligence (AI) and machine learning (ML) solutions shine.

Oracle Cloud Infrastructure (OCI) Anomaly Detection is an AI service that provides real-time and batch anomaly detection. The service helps developers more easily build business-specific anomaly detection models that flag critical incidents, resulting in faster time to detection and resolution. Proprietary algorithms detect anomalies earlier with fewer false alarms. Intelligent data processing techniques account for errors and imperfections in real-world input data, automatically identifying and fixing data quality issues. Custom-trained AI models come with APIs that let developers upload their business-specific data, so they can detect anomalies using the stored model. These features make highly accurate, custom-trained anomaly detection models accessible to everyone, even companies without data science expertise. Easy access to open-source technologies also expands usage of OCI Anomaly Detection’s models.

Let’s look at OCI technologies in action with a banking customer.

OCI in action

Certegy provides banking authorization and risk management of $43 billion in check processing for more than 4,000 retailers in more than 300,000 locations around the world. Certegy uses Oracle Analytics and Autonomous Database with ML and spatial analysis to help businesses minimize losses and prevent fraud.

“By using the Oracle Analytics and Autonomous Data Warehouse, our goal is to apply machine learning and spatial analysis to better track check cashing behavior that mitigates risk and prevents fraud in real time to help businesses and consumers more confidently engage in commerce,” said Eric Probst, senior manager of fraud analytics at Certegy.

OCI fuels financial innovation through simplifying risk management, allowing banking institutions to harness data for insights that drive value. Oracle takes an end-to-end view of your bank’s value chain.

Want to know more?

To learn how banks can accelerate the account opening and onboarding process, improve abandonment rates, and meet modern day customer expectations, watch the replay of the Detecting Anomalies on Credit Card Transactions webinar.

This article is part of a series on cloud-enabled innovation in the financial services industry. Check out related editorials on account opening and onboarding and loan origination.