Abstract
A fuzzy sets based structure for classification systems is proposed in this
paper. The basic idea is to use "windows" to represent the constraints on
the possible values variables may take. The formalism is very simple,
however, this simplicity makes it attractive in the development of
knowledge based systems. The most salient features of this structure
include the possibility of developing composite solutions, searching for
lacunae, and creating a case based representation of knowledge with an
avenue for modeling learning. (Autor)