Mardi | 2019-11-05
Salle des thèses 16h-17h20
Financial institutions increasingly rely on predictive machine learning models to detect fraudulent transactions. Two main challenges when building a supervised tool for fraud detection are the imbalance or skewness of the data and the various costs for different types of miscalissification. We discuss techniques to solve the imbalance issue and present a cost-sensitive logistic regression algorithm.