Interpolative reasoning methods is a reasoning technique that is designed to deal with reasoning in sparse rule-based systems. This paper proposed a fuzzy interpolative reasoning method by using a like - gravity - center of fuzzy sets whose shapes are trapezoidal. This method allows the conditions appearing in the antecedent part and the consequence of the rules to be represented by trapezoidal fuzzy numbers. The work steps of the presented method are first constructing a new inference rule by manipulating two given adjacent rules and next by exploiting similarity information to convert the derived inference result into the conclusion. In this process, we use scale and move rate transformation operation to support such reasoning.
展开▼
