Towards Evolving Parametric Fuzzy Classifiers Using a Virtual Sample Generation Approach
Holger Hähnel, Arne-Jens Hempel, Gernot Herbst · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2015
Evolving classification models are designed to solve online tasks with demands restricting computational power and memory.The present paper proposes an evolving version of an established fuzzy classification approach based on fuzzy pattern classes.The approach incorporates a novel type of virtual sample generation.It creates examples from given parametric model information and thus inverts the classifier's batch learning algorithm.In an evolving environment, virtual examples and real learning data are combined for on-line learning.The main advantage of this approach is that the original learning process retains its applicability while memory demands are reduced significantly.Academic examples demonstrate the feasibility.