Fuzzy Classification of Nonconvex Data-Inherent Structures

Abstract

The present work dedicates itself to the aggregation of nonconvex data-inherent structures into fuzzy classes. A key feature of this aggregation is its conduction within a closed fuzzy classification framework, being built around a single, generic type of a convex membership function. After a short elaboration concerning this essential building block a novel automated, data-driven design strategy to aggregate complex (nonconvex) data-inherent structures is introduced. The whole aggregation process will be illustrated with the help of an example. © Springer-Verlag Berlin Heidelberg 2010.