Using a genetic algorithm-based classifier system for modeling auditor decision behavior in a fraud setting

This paper addresses a classification problem involving the decisions of Defense Contractor Audit Agency (DCAA) auditors when they are estimating the likelihood of fraud by contractors developing bids for government contracts. The objective of the study is to investigate if this decision involves non-algebraic processes associated with a posited simultaneous decision model or algebraic processes posited by sequential decision processes. We propose that in classification decision models involving simultaneous processing, genetic algorithms represent an innovative heuristic approach, which may produce improved models when compared to traditional mathematical approaches. © 1998 John Wiley & Sons, Ltd.