In credit scoring, low-default portfolios are those for which very little default history exists. This makes it problematic for financial institutions to estimate a reliable probability of a customer defaulting on a loan. In this talk the suitability of one-class classification algorithms as a solution to the low-default portfolio problem are evaluated. The performance of one-class classification algorithms is compared with the performance of supervised two-class classification algorithms. This talk also investigates the suitability of oversampling, which is a common approach to dealing with low-default portfolios. A second topic covered in this talk is a framework to generate artificial data that can simulate population drift in credit scoring. Population drift undermines the performance of supervised classification models by contravening the expectation that the characteristics describing the customer remain stationary over time. To ensure that our framework is sufficiently grounded in reality, data distributions are generated using a troika of sources: demographic information from the Central Statistics Office Ireland; housing statistics published by the Irish Government; and we develop a profile of loan defaulters using a recent Moody’s report ("What Drives Irish Mortgage Borrowers to Default"). Through user controlled settings the interaction between features can be adjusted over time to simulate population drift.