Fuzzy Cognitive Maps
Advances in Theory, Methodologies, Tools and Applications
Buch in deiner Nähe kaufen
Beschreibung
The theory of cognitive maps was developed in 1976. Its main aim was the representation of (causal) relationships among “concepts” also known as “factors” or “nodes”. Concepts could be assigned values. Causal relationships between two concepts could be of three types: positive, negative or neutral. Increase in the value of a concept would yield a corresponding positive or negative increase at the concepts connected to it via relationships.
In 1986 Bart Kosko introduced the notion of fuzziness to cognitive maps and created the theory of Fuzzy Cognitive Maps (FCMs). The relationship between two concepts in (FCMs) can take a value in the interval [-1,1]. This relationship value is called “weight”.
For the last twenty years extensive research in the theory of FCMs has been performed that provided major improvements and enhancements in its theoretical underpinning. New methodologies and approaches have been developed. FCMs have also been applied to many different sectors. New software tools have been developed that automate FCM creation and management.
The aim of this book is to present recent advances and state of the art in FCM theory, methodologies, applications and tools that exist to date scattered in journal papers, in a concrete and integrated manner.
This important edited volume is the first such book ever published on fuzzy cognitive maps (FCMs). Professor Michael Glykas has done an exceptional job in bringing together and editing its seventeen chapters. The volume appears nearly a quarter century after my original article “Fuzzy Cognitive Maps” appeared in the International Journal of Man-Machine Studies in 1986. The volume accordingly reflects many years of research effort in the development of FCM theory and applications—and portends many more decades of FCM research and applications to come. FCMs are fuzzy feedback models of causality. They combine aspects of fuzzy logic, neural networks, semantic networks, expert systems, and nonlinear dynamical systems. That rich structure endows FCMs with their own complexity and lets them apply to a wide range of problems in engineering and in the soft and hard sciences. Their partial edge connections allow a user to directly represent causality as a matter of degree and to learn new edge strengths from training data. Their directed graph structure allows forward or what-if inferencing. FCM cycles or feedback paths allow for complex nonlinear dynamics. Control of FCM nonlinear dynamics can in many cases let the user encode and decode concept patterns as fixed-point attractors or limit cycles or perhaps as more exotic dynamical equilibria. These global equilibrium patterns are often “hidden” in the nonlinear dynamics. The user will not likely see these global patterns by simply inspecting the local causal edges or nodes of large FCMs.
Contains the theory of fuzzy cognitive mapping and for the first time a methodology of constructing and linking fuzzy cognitive maps to be used in management and business process reengineering for performance measurement purposes Contains a software tool that can be used for designing and simulating FCMs that can be linked and simulated as an integrated project Written by experts in the field
Autor*in
Michael Glykas
Themen in »Fuzzy Cognitive Maps«
Fuzzy Cognitive Maps Performance Measurement artificial intelligence calculus complex systems computer vision control decision tree expert system fuzziness fuzzy knowledge knowledge management modeling simulation