One of the techniques for approximate reasoning is the revision principles, such as the linear revising and the semantic revising methods. These methods can not give an interpretable conclusion and can not be applied in every case: they have different strict conditions for the antecedent and the consequent sets. In order to solve this problem a general revision principle method is proposed in this paper, which can be applied to arbitrary fuzzy sets. The new method can be introduced as the extension of the former revision principle methods and a special case of the general fuzzy rule interpolation method. These two significant parts of fuzzy reasoning are discussed from one viewpoint based on the proposed method.
展开▼