IFCM:Fuzzy clustering for rule extraction of interval Type-2 fuzzy logic system
Weibin Zhang, Wenjiang Liu · 2007
Compared with the traditional type-1 fuzzy logic system, type-2 fuzzy logic systems (T2FLS) are suitable to handle the situations where a great deal of uncertainty are present. However, how to extract fuzzy rules automatically from input/output data is still an important issue because sometimes human experts can not get valid rules from unknown systems. Fuzzy c-means clustering (FCM) is one of algorithms used frequently to extract rules from type-1 fuzzy logic system, but its application is merely limited to dots set. This paper introduces an enhanced clustering algorithm, called the interval fuzzy c-means clustering (IFCM), which is adequate to deal with interval sets. Moreover, it is shown that the proposed IFCM algorithm can be used to extract fuzzy rules from interval type-2 fuzzy logic system. Simulation results are included in the end to show the validity of IFCM.