Presented By O’Reilly and Cloudera
Make Data Work
March 5–6, 2018: Training
March 6–8, 2018: Tutorials & Conference
San Jose, CA

Detecting retail fraud with data wrangling and machine learning

Matt Derda (Trifacta), Harrison Lynch (Consensus Corporation)
2:40pm3:20pm Thursday, March 8, 2018
Average rating: **...
(2.00, 1 rating)

What you'll learn

  • Learn how Consensus leverages the combined power of data wrangling and machine learning to more efficiently identify and reduce retail fraud


Fraud is a top concern for retailers, especially when it involves costly devices like smartphones or televisions. For every one device “purchased” with a fraudulent credit card, retailers must sell five more in order to recoup those losses. Consensus Corporation has built a fraud detection model as part of its Revenue Cloud on the Connected Commerce platform in order to alert retailers of potential activity before it occurs. The more efficiently Consensus can update this model with new retail and consumer data, the more money it can save retailers—even just a small percentage of efficiency gains could translate to millions saved.

Matt Derda and Harrison Lynch explain how Consensus leverages the combined power of data wrangling and machine learning to more efficiently identify and reduce retail fraud and how adopting data wrangling technology has helped Trifacta reduce time spent data wrangling from six weeks to one week.

Photo of Matt Derda

Matt Derda


Matt Derda leads customer marketing at Trifacta, where he focuses on helping customers communicate the value they’re generating from initiatives to modernize data management, analytics and machine learning. Matt’s previous experience as a data wrangler at PepsiCo means that he is well-versed in the challenges of getting data ready for various analytics initiatives. He uses this experience to establish strong relationships with Trifacta customers and help them promote their innovative work to internal and external audiences.

Photo of Harrison  Lynch

Harrison Lynch

Consensus Corporation

Harrison Lynch is the Senior Director of Product Management for the Consensus Marketplace Cloud and Risk Cloud products. Ever the Digital Sheriff, even when deep in the midst of a re-platforming and developing an API product offering, he keeps himself awake at night pondering how to prevent identity thieves from stealing iPhones and how to manage multiple Product Managers and dev teams. Harrison has spent the past fifteen years of his career in the niche world of integrating cell phone carrier systems to web applications, while trying to stop fraudsters. He loves a good nerd fight, and has never seen a Product Management blog that he won’t dive into.

Harrison is a 1995 graduate of the University of Oregon.