AI Approaches to Fraud Detection and Risk Management
Papers from the AAAI Workshop
Tom Fawcett, Chair
Fraud detection and risk management involve monitoring the behavior of populations of users in order to estimate, detect or avoid undesirable behavior. Undesirable behavior is a broad term including delinquency, fraud, intrusion and account defaulting.
This workshop brought together researchers in these areas to discuss approaches and experiences in dealing with the critical issues of arge volumes of data, highly skewed distributions, changing distributions, widely varying error costs, and costs changing over time, adaptation of undesirable behavior to detection techniques, changing patterns of legitimate behavior, and social issues (privacy, discrimination, "redlining").