Decision Making at Scale: Signifyd ML Models - Tech Stack Best Practices - FLOW Summit 2021

Machine learning has become an integral part of any modern-day fraud strategy - But what’s really happening under the hood?
The rising importance of ecommerce to retail businesses and rapidly changing fraud landscape has required organizations to seek more sophisticated methods of fraud detection, such as machine learning, to learn, adapt, and uncover emerging fraud patterns quickly. Beyond just superior detection, these tools have also proven their value in drastically reducing manual labor for fraud teams and can scale to manage even the most impressive peak sales periods with ease.
But are all machine learning models created equal? How can you trust decision that are being made for your business? How do you know which solution is right for you?
Join Swami Vaithi, SVP of Decision Science at Signifyd as he deep dives into Signifyd’s own machine learning philosophy. Tune in to learn more about:
• Transforming information into useful machine learning data
• The approach Signifyd takes to feature engineering
• Algorithms best suited for classification problems
• Model validation and incorporating feedback
• How you can compare and choose the best machine learning solution for your business
00:00 Introduction by Swami Vaithi, SVP of Decision Science, Signifyd
01:21 Signifyd’s holistic approach
03:25 Commerce Protection Platform
10:29 Signifyd’s approach to feature engineering
15:57 Algorithems used by Signifyd
17:16 Model creation life cycle
19:29 Explanation of the continuous learning
Stay up-to-date with FLOW Summit 2024 on demand:
www.signifyd.com/flow-summit-...