The application of generalised fuzzy rules to machine learning and automated knowledge discovery

Notions of generalised fuzzy conditional and equivalence rules relative to a combination function are introduced and a framework for reasoning with su ch rules is described. The applicability of this framework to machine learn ing and knowledge discovery problems is demonstrated. Methods for the autom atic generation of two particular types of generalised rule are proposed. T he two rule forms considered are rules with weighted AND/OR combination fun ctions, as suggested by Zimmerman and Zysno, and evidential logic equivalen ce rules as defined by Baldwin. The process of generating rule bases is div ided into the problem of generating fuzzy sets from data and that of findin g combination functions to optimise the performance of the system given the se fuzzy sets. For the former problem a mass assignment based approach is a dopted and for the latter semantic discrimination analysis is used in conju nction with customised optimisation algorithms. The potential of rule bases of both forms is illustrated by their application to a number of model and real world machine learning problems.