CLASSIFYING THROUGH A FUZZY ALGEBRAIC STRUCTURE

The properties of a fuzzy algebraic structure are investigated and the existence and uniqueness of the operators for the elements of the str ucture is formally proved. Afterwards the problem of generating approp riate classifications in a universe of discourse in which the characte ristics of the elements are defined through attributes whose values ar e linguistic labels is analyzed and a method based on the fuzzy struct ure is introduced. A simple example is used to show how the new method works and results are compared to the ones given on the same data set by better-known clustering algorithm, in order to prove feasibility. Finally, an application of the fuzzy clustering in the context of a re al medical diagnosis will be shown and main features of the method dis cussed.