Multiple algorithms for fraud detection

This paper describes an application of Case-Based Reasoning to the problem of reducing the number of final-line fraud investigations in the credit app roval process. The performance of a suite of algorithms, which are applied in combination to determine a diagnosis from a set of retrieved cases, is r eported. An adaptive diagnosis algorithm combining several neighbourhood-ba sed and probabilistic algorithms was found to have the best performance, an d these results indicate that an adaptive solution can provide fraud filter ing and case ordering functions for reducing the number of final-line fraud investigations necessary. (C) 2000 Elsevier Science B.V. All rights reserv ed.