Fuzzy multi-stage inference apparatus.

摘要

In a method (claim 1), an inference is conducted by use of fuzzy knowledge and knowledge of experts can be directly written in inference rules. In a method (claims 2 to 4), it is possible to use ambiguous language expressions of the user and the user can freely describe flexible knowledge, although ambiguous, in a knowledge base. In a method (claims 5 to 7), the user clearly specifies execution priority orders or conviction degrees of inference rules such that even when an antecedent contains a confirmative or determinant action, a contention between rules is prevented to execute an optimal inference rule in any situation. In a method (claims 8 to 15), an intermediate result of a fuzzy inference is passed in a form of a real number or a fuzzy label to the next inference to achieve a fuzzy multi-stage inference, thereby conducting an inference more similar to the human inference as compared with a single-stage inference. In a method (claims 16 to 18), a multi-stage inference conducting a backtrack operation based on fuzzy knowledge is implemented to easily reflect user's ambiguous knowledge onto inference rules. In a method (claims 19 to 26), a new infernece method associated with a best value retrieval function is provided wherein, in contrast to the conventional method in which a fuzzy inference and a binary logic inference are continuously achieved from an upper-level retrieval or search point to a lower-level retrieval point in a one-directional manner, the best value retrieval function is employed, thereby easily obtaining an optimal solution. In a method (claims 27 and 28), the user can describe inference rules without perceiving any difficulty of a description of the termination instruction to terminate an inference at a satisfactory inference stage, thereby configuring a system which matches with the human fuzziness and which can be easily operated by the user.