New Interpolative Reasoning in Sparse Fuzzy Rule Base

摘要

Fuzzy reasoning is a very important part in intelligent system, when rule base is sparse and the observation is in the "gap" between two neighboring antecedents of rules, we cannot get any reasoning consequence by traditional CRI method. Fuzzy reasoning is essentially an interpolator. Koczy and Hirota proposed a linear interpolative reasoning method called KH linear interpolative reasoning method. But it is also difficult to preserve convexity and normality by using KH linear interpolative reasoning method. Based on the analysis of KH linear interpolative reasoning method, in order to get better reasoning consequence under the condition of sparse rules, we propose a new linear interpolative reasoning method that can preserve the convexity and normality of the reasoning consequence better. It devotes a useful tool for fuzzy reasoning in intelligent systems.