Abstract: Fuzzy class can be obtained via fuzzy relevance degrees approach. The relationships among the attributes can be described by fuzzy rule s enclosed in the fuzzy class. And the membership functions, which are used to calculate the membership degree of subclass and superclass, and the members hip degree of object and class, are defined according to fuzzy relevance degrees and fuzzy range. Thereby, the problem of imprecise  and uncertain information acquirement can be settled. As a result, this approach can be applied to model ing the imprecise requirement.

Key words: fuzzy relevance degrees, fuzzy class, imprecise requi rement