Título Dealing with Complexity in Large Scale and Structured Fuzzy Systems
Autores GARCÍA ALONSO, CARLOS, Ma, J, Yin, YL, Yu, J, Zhou, SG, GARCÍA ALONSO, CARLOS
Publicación externa No
Medio Fifth International Conference On Fuzzy Systems And Knowledge Discovery, Vol 3, Proceedings
Alcance Proceedings Paper
Naturaleza Científica
Ámbito Internacional
Fecha de publicacion 01/01/2008
ISI 000264269100055
DOI 10.1109/FSKD.2008.342
Abstract Fuzzy inference engines must always deal with the complexity involved in an exponentially increasing number of rules. Sometimes in complex problems, it is difficult to have expert knowledge at one's disposal to design the whole rule set. Nevertheless, experts can guide the rule design by defining the variables involved and giving guidelines about their behavior. A dependence relationship (DR) is a set of rules defined by a group of related inputs and outputs. In order to make the design and evaluation of DRs automatic, two properties called type and intensity are introduced. The DR type identifies the output membership functions shifting the neutral selection to the right or to the left. The DR intensity qualifies the final output membership function selection admitting the existence of nuances in rule fulfillment. Applying these properties, DR rules can be automatically designed and appropriately interpreted by the fuzzy inference engine in complex systems.
Miembros de la Universidad Loyola

Change your preferences Gestionar cookies