Fuzzy mechanisms for qualitative causal relations

Abstract

Fuzzy systems main asset over competing techniques has always been the capability to model expert qualitative knowledge. However, probably due to the limited scientific appeal, there has always been a scientific trend to disregard this simple but effective asset in favor of more hard-mathematical aspects of fuzzy systems. This can prove to be a mistake, especially when approaching qualitative real world dynamic systems, like, for instance, Social, Economical or Political Systems. Such systems are composed of a number of dynamic concepts or actors which are interrelated in complex ways usually including feedback links that propagate influences in complicated chains. Axelrod [1] introduced Cognitive Maps (CMs) as a way to represent and analyze the structure of those systems, but techniques that allow simulating the evolution of cognitive maps through time, what one could call Dynamic Cognitive Maps (DCM), were not available or had serious limitations during more than two decades [5], [9]. Fuzzy sets should have been regarded as the ideal 'tool' when considering modeling such systems. However, proper qualitative modeling was consecutively disregarded even when fuzzy systems were used by Kosko to approach the problem (Fuzzy Cognitive Maps) [3], [4], [5], [11], [12], [13]. Rule Based Fuzzy Cognitive Maps (RB-FCM) were introduced has a qualitative technique to solve the limitations of previous approaches to this problem. They can be used as a tool by non-engineers and/or non-mathematicians since they eliminate the need for complex mathematical knowledge when modeling dynamic qualitative systems. © 2009 Springer-Verlag Berlin Heidelberg.