Abstract
In everyday life we sometimes encounter the vague statements whose truth status are neither true nor false. For developing the reasoning method using such vague information, the technique of Truth Qualification proposed by Zadeh is very important. This paper proposes new methods of Truth Qualification and its converse problem. We deal with the case in which the truth status associated with fuzzy propositions are given by numerical truth values rather than fuzzy truth values. Furthermore, with the use of the present method and the implication rule in Lukasiewicz logic, we formulate fuzzy reasoning methods for the cases corresponding to modus ponens and modus tollens of classical logic. The results obtained are satisfactory in view of normative criteria to be possessed by fuzzy reasoning. Also the present method is so simple that it is easy to give the intuitive interpretation to each process in fuzzy reasoning.
