Fraud & Risk

An Executive's Guide to the Evolution of Fraud Detection

Julian Wong - DataVisor


Fraud management systems are typically composed of a variety of defenses, with rules engines and supervised machine learning models anchoring many eCommerce companies' fraud detection strategies. After reviewing how those systems work and the benefits and drawbacks of each, this whitepaper examines a newer technology called unsupervised analytics. Two main categories of unsupervised algorithms and how they work are described, along with a real-world example of a promotional abuse attack to help illustrate how unsupervised analytics can handle previously unrecognized forms of fraud. The paper concludes with four takeaways on unsupervised analytics for fraud and risk leaders.

An Executives Guide to the Evolution of Fraud Detection

Download Now