Proposal of the intellectual evaluation using fuzzy memory-based reasoning and fuzzy measure

The method of "state interpretation by fuzzy memory-based reasoning" and the identification method of "fuzzy measure and scenario integrated evaluation scale" are proposed to construct the intellectual evaluation considering the context. Conventional methods evaluate the object by interpreting the state according to fragmentary knowledge acquired from experts. The interpretation, however, often depends on contexts, that is, other states that exist at the same time. The problem is that the evaluation result has been ambiguous in the usual knowledge representation because the interpretation of the state does not change even if the context changes. The method of interpreting the state by fuzzy memory-based reasoning proposed eliminates the fuzziness by using the bottom up and the top down processing mutually to reorganize the information so that the state suits its strategy, or its context. The method of identifying fuzzy measure proposed clarifies indirect intentions, or nonlinear relation between the state and the strategy based on the appropriate context information, which eliminates the ambiguity included in the state evaluation. Furthermore, these methods are applied to the adjustment of the cooling capability in a vending machine. >