稀疏规则条件下的线性插值推理研究

模糊推理本质上就是某种插值方法．但在稀疏规则库的条件下，当输入的事实落入规则“空隙”时，采用传统的CRI方法是得不到任何推理结果的．而采用KH线性插值推理也存在着难以保证推理结果的凸性和正规性等问题．在分析了Koczy和Hirota提出的线性插值推理方法的基础上，本文提出了一个新的线性插值推理的方法，该方法能很好地保证推理结果的凸性和正规性，这为智能系统中的模糊推理提供了一个十分有用的工具．

Reasearch on Linear Interpolative Reasoning for the Irregular Fuzzy Rule

Fuzzy reasoning method is equal to some interpolative method. When rule base is sparse, we can not get any reasoning result by traditional CRI method for an observation is in the gap between two neighboring antecedents. It is also difficult to keep convexity and normality using KH linear interpolative reasoning method. After analyzing the principle of interpolative reasoning proposed by Koczy and Hirota, we propose a new interpolative reasoning method that can keep the convexity and normality of the reasoning consequence. It devotes a useful tool for fuzzy reasoning in intelligent systems.