Fraud & Risk
Using Machine Learning to Combat Fraud at Scale
Deliveroo, a British restaurant delivery company, migrated from a rules-based fraud detection system to a machine learning system. This presentation examines how the company moved past the so-called "cold start" problem and how the company opted to handle customer inquiries in lieu of a manual review process. The discussion concludes with a look at the business impact resulting from the migration and how the company plans on leveraging other features and use cases on Ravelin's platform, such as graph networks.