Learn about the latest trends and insights in the world of online fraud from our analysis of over 4 billion accounts from some of the largest internet properties and financial services in the world.
DataVisor's VP of Engineering David Ting discusses how DataVisor optimizes its AWS stack with spot fleets, dynamic instance launches, & RT cost tracking, all while protecting over 4B users from fraud at the AWS Summit Anaheim 2018.
As attacks grow in scale and velocity, businesses are forced to evolve their fraud detection methods from manual detection involving blacklists and rule engines to machine learning algorithms that can detect known and emerging types of fraud. This article highlights why existing fraud detection methods have limitations and more importantly a few reasons why unsupervised machine learning is gaining traction.
DataVisor's Yuhao Zheng and Boduo Li share advanced techniques for managing thousands of spark workers to analyze billions of events at a time, including clustering workers and automated, optimized management of DataVisor's spark infrastructure.
DataVisor's Ting-Fang Yen and Arthur Meng present a novel deep learning technique for scalable online fraud detection among billions of users.
Does the fact that UML doesn’t require labels mean that there is no benefit at all to labels? If label data exists already, how can it be used to improve UML detection results? In this article we discuss how labels can be effectively used in UML detection, even if they are not required.
Introduction There are many technical articles that describe supervised and unsupervised machine learning methods. In this guide, we will explain a few high level differences when it comes to choosing between the two. Comparison 1: [...]
The DataVisor Online Fraud Report took a look at our base of more than one billion users across 172+ countries in the world. Using this massive amount of data, we were able to identify some of the favorite tools and attack techniques that online criminals from around the globe favor when doing their dirty work.
As mentioned in my previous articles, traditional rule-based transaction monitoring systems (TMS) have architectural limitations which make them prone to false positives and false negatives: Naive rules create a plague of false positives that are [...]