arrow left facebook twitter linkedin medium menu play circle
June 2, 2021 - Claire Zhou

Meet the Next-Gen Proactive Fraud Prevention Technology

Fraud continues to evolve at lightspeed, which means fraud prevention technology must do the same to keep up. Now, with DataVisor’s comprehensive suite of fraud prevention tools and technology, it’s no longer a matter of staying one step behind fraudsters, but rather staying in step alongside them — and in many cases, being able to stop fraud at the gate and eliminate the need for damage control.

DataVisor’s next-gen proactive fraud prevention technology is making real-time detection and prevention a reality for many businesses and a possibility for others. Here’s a closer look inside DataVisor’s sophisticated suite of tools and tech.

Unsupervised Machine Learning

Unsupervised machine learning (UML) leverages the power of machine learning but doesn’t require the use of labeled input data. Rather, it reviews millions of data points in real time, both structured and unstructured, to find patterns and connections between data that may indicate fraud. This type of fraud detection technology can be extremely useful in finding unknown fraud patterns that models haven’t yet been trained to identify.

What’s more, because UML doesn’t rely on extensive data training, fraud models don’t decay as quickly and will continue to detect fraud at a high accuracy even months after deployment. The models are self-adapting and don’t require constant retuning. This can save time and resources in fraud departments without sacrificing performance and results.

Device Intelligence

Fraud occurs across a variety of channels and devices, some of which are harder to detect than others. For example, devices that use emulators or are jailbroken can be hard to trace, and therefore difficult to link to fraud patterns.

DataVisor’s Device Intelligence aims to eliminate this risk by assigning a device fingerprint to every device and protect companies against bot attacks, app cloners, and emulators. Even when fraudsters change device parameters, they cannot change their DataVisor-assigned device ID. 

Feature Platform

Historically, one of the most time-consuming and costly parts of developing models is the investment of developing features. This isn’t the case with DataVisor’s Feature Engineering, which offers robust features right out of the box. This means companies no longer have to build fraud detection features from scratch and can still get the level of customization they need to support their fraud programs. The core of DataVisor’s feature platform is its data enrichment and advanced feature engineering. 

DataVisor’s Feature Platform is highly scalable and can handle years of data for large institutions and process massive amounts of data in real time — 10,000+ QPS with only 10-50 ms latency. It also supports backtesting and forward testing on extensive historical data to achieve optimal performance.

Knowledge Graph

Adding context to flagged activities, DataVisor’s Knowledge Graph creates a visual linkage among entities so teams can understand the reasons behind the findings. The graph is automatically generated based on millions of data points from omnichannel data. Teams can visualize the patterns, investigate cases with a single click, and take bulk actions with entities or groups directly from the graph interface.

Want to meet our next-gen proactive fraud prevention technology in person? Learn more about DataVisor’s technology. 

Photo of Claire Zhou
about Claire Zhou
Claire is a Product Marketing Manager at DataVisor with over 5 years of marketing experience in security and fin-tech. She is passionate about empowering enterprise customers with AI-based solutions. Her expertise spans data analytics, cybersecurity, and fraud prevention. Claire has an MBA from UCLA.
Photo of Claire Zhou
about Claire Zhou
Claire is a Product Marketing Manager at DataVisor with over 5 years of marketing experience in security and fin-tech. She is passionate about empowering enterprise customers with AI-based solutions. Her expertise spans data analytics, cybersecurity, and fraud prevention. Claire has an MBA from UCLA.