Financial Classification Using an Artificial Immune System
Anthony Brabazon (University College Dublin, Ireland), Alice Delahunty (University College Dublin, Ireland), Dennis O’Callaghan (University College Dublin, Ireland), Peter Keenan (University College Dublin, Ireland) and Michael O’Neill (University of Limerick, Ireland)
Copyright: © 2008
Recent years have seen a dramatic increase in the application of biologically-inspired algorithms to business problems. Applications of neural networks and evolutionary algorithms have become common. However, as yet there have been few applications of artificial immune systems (AIS), algorithms that are inspired by the workings of the natural immune system. The natural immune system can be considered as a distributed, self-organizing, classification system that operates in a dynamic environment. The mechanisms of natural immune systems, including their ability to distinguish between self and non-self, provides a rich metaphorical inspiration for the design of pattern-recognition algorithms. This chapter introduces AIS and provides an example of how an immune algorithm can be used to develop a classification system for predicting corporate failure. The developed system displays good classification accuracy out-of-sample, up to two years prior to failure.