A new approach to connective generation in the framework of expertsystems using fuzzy logic

摘要

A technique for modeling uncertainty in expert systems isnpresented. Operators are defined using linguistic terms to avoid anynnumerical representation. These operators consider linguistic term setnordering, constraints that are the counterpart of properties fulfillednby triangular norms in fuzzy logic, and additional restrictions creatednby the expert's procedure to combining certainty. The method avoids thenusual problems that arise in other treatments of certainty linguisticnterms, where they are represented as fuzzy numbers or fuzzy truthnlabels. One of the most significant problems is the lack of consensus innthe representation of each term by a group of experts, due to thennecessity of representing the terms in a pseudonumerical scale