Article ID: | iaor2000573 |
Country: | Hungary |
Volume: | XXIX |
Issue: | 1/2 |
Start Page Number: | 7 |
End Page Number: | 28 |
Publication Date: | Jan 1998 |
Journal: | Szigma |
Authors: | Borgulya Istvn |
Keywords: | fuzzy sets |
Two heuristic fuzzy algorithms are shown in this paper: a fuzzy classification and a fuzzy cluster algorithm. The algorithms classify multiple-criteria fuzzy or crisp alternatives and the task of the classification is traced back to ranking of alternatives, as well as to learning weight numbers of criteria similarly to neural networks. The algorithm is suitable to separate classes with relatively high intrinsic scatter.