A Study On The Interaction In Fuzzy Production Rule Reasoning

Knowledge reasoning is one of the most important processes of machine thinking. The uncertainty of knowledge and information is mostly aroused by ambiguity. It leads to the study of fuzzy reasoning extraordinarily important. Production rules is a commonly used knowledge representation pattern, it has been widely used in many successful expert systems. So the study of fuzzy production rule reasoning has become a hotspot in the field of artificial intelligence. When fuzzy production rules are used to approximate reasoning, interaction exists among rules that have the same consequent. In order to model and handle this interaction, this paper proposes to use a non-additive nonnegative set function to replace the weights assigned to rules having the same consequent, employ Sugeno fuzzy integral to calculate the result of production rules which have the same consequence and establish the mathematic model to determine the non-additive set function from data. The simulation experiment shows that handling interaction in fuzzy production rule reasoning in this way can lead to a good understanding of the rules base an improvement of classification accuracy and reasoning capability.