Effect of rule weights in fuzzy rule-based classification systems

This paper examines the effect of rule weights in fuzzy rule-based classifi cation systems. Each fuzzy IF-THEN rule in oar classification system has an tecedent linguistic values and a single consequent class. We use a fuzzy re asoning method based on a single winner rule in the classification phase. T he winner rule for a new pattern is the fuzzy IF-THEN rule that has the max imum compatibility grade with the new pattern. When we use fuzzy IF-THEN ru les with certainty grades (i.e., rule weights), the winner is determined as the rule with the maximum product of the compatibility grade and the certa inty grade. In this paper, the effect of rule weights is illustrated by dra wing classification boundaries using fuzzy IF-THEN rules with/without certa inty grades. It is also shown that certainty grades play an important role when a fuzzy rule-based classification system is a mixture of general rules and specific rules. Through computer simulations, we show that comprehensi ble fuzzy rule-based systems with high classification performance can be de signed without modifying the membership functions of antecedent linguistic values when we use fuzzy IF-THEN rules with certainty grades.