
Brooks Bourland
Head of Global Digital Fraud Data Science
TransUnion
Operationalizing ML/AI in the Fraud Mitigation Space
- Fraud prevention represents a dynamic and challenging area for data science and analytics. Building and assessing performance of machine learning solutions often is the tip of the iceberg in terms of the having an impact in preventing fraud. Deployment within the real-time architecture and decision rules constitute the bulk of the work. Given this context it is critical to understand how to evaluate the performance of individual machine learning models, and in developing comprehensive, data driven, fraud prevention engines. This discussion will seek to provide an overview of how we 1) assess value of individual solutions, 2) operationalize multiple machine learning solutions, and 3) using machine learning to create optimal decision rules.