Integrated case-based neural network approach to problem solving

Abstract

Case-based reasoning can be a particularly useful problem solving strategy when combined with other artificial intelligence reasoning paradigms or with some other computational problem solving method. An approach is presented in which the machine learning capabilities of an artificial neural network are used to enhance the reuse of past experience in the case-based reasoning cycle. This approach has been found to be effective in the application of case-based reasoning to forecasting. © 1999 Springer-Verlag Berlin Heidelberg.