Using single-antecedent fuzzy rules in fuzzy knowledge map
Sebastian W. Khor, Хан, K.W. Wong · 2005
Conventional fuzzy inference methodology relies on the mapping of the input spaces to the output space by partitioning the spaces with membership functions. In cases where there are more than one input variables, an intersection of memberships is adopted by aggregating these regions. This strategy yields an exponential growth in the number of rules as inputs are added to the system, quickly reducing performance to unacceptable levels. We present a methodology that allows the use of single antecedent fuzzy rules to approximate a class of problems in the Fuzzy Knowledge Map- a knowledge representation framework developed by us.