Evaluating medical decision making heuristics and other business heuristics with neural networks

Abstract

Heuristics are an efficient means for solving complex and also partial information business problems. Unfortunately, the development of new heuristics and the evaluation of existing heuristics is a labor intensive process. Neural networks provide a fast and reliable method for evaluation of new heuristics against existing heuristics and the optimization of new heuristics when no prior heuristic exists. This chapter describes a methodology for utilizing neural networks as a heuristic evaluation mechanism and discusses how existing research has been utilized (possibly unintentionally) in the development or evaluation of new heuristics. © 2008 Springer-Verlag Berlin Heidelberg.