The article concentrates on selected aspects of knowledge representation. It is considered how fuzzy sets viewed as linguistic labels (information granules) determine a representation frame. the representation scheme is discussed with respect to its perception level and a scope of individual labels. Both the features can be handled by fuzzy sets and easily modeled by changing associated parameters of corresponding membership functions. Problems of prototype determination in the original space (in which the objects are characterized) as well as in the space of linguistic labels are studied. It is argued that the space of grades of membership functions is more suitable to capture a notion of the prototype. Different aspects of matching are also brought into account.
Discussion(0)
No comments yet. Be the first to comment.