An adaptive fuzzy system for control and clustering of arbitrary data patterns

Abstract

A modular, unsupervised neural network architecture is described. It can be used for data clustering and classification. The adaptive fuzzy leader clustering (AFLC) architecture is a hybrid neural-fuzzy system which learns online in a stable and efficient manner. The system consists of a fuzzy k-means learning rule embedded within a control structure similar to that found in the adaptive resonance theory (ART-1) network. AFLC adaptively clusters analog inputs into classes without prior knowledge of the entire data set or of the number of clusters present in the data. The classification of an input takes place in a two-stage process; a simple competitive stage and a euclidean metric comparison stage. The AFLC algorithm and its operating characteristics are described. The algorithm is compared to an adaptive Bayesian classifier for some real data.<>