T. Frank et al., COMPARATIVE-ANALYSIS OF FUZZY ART AND ART-2A NETWORK CLUSTERING PERFORMANCE, IEEE transactions on neural networks, 9(3), 1998, pp. 544-559
Adaptive resonance theory (ART) describes a family of self-organizing
neural networks, capable of clustering arbitrary sequences of input pa
tterns into stable recognition codes. Many different types of ART-netw
orks have been developed to improve clustering capabilities. In this p
aper we compare clustering performance of different types of ART-netwo
rks: Fuzzy ART, ART 2A with and without complement encoded input patte
rns, and an Euclidean ART 2A-variation. All types are tested with two-
and high-dimensional input patterns in order to illustrate general cap
abilities and characteristics in different system environments. Based
on our simulation results, Fuzzy ART seems to be less appropriate when
ever input signals are corrupted by addititional noise, while ART 2A-t
ype networks keep stable in all inspected environments. Together with
other examined features, ART-architectures suited for particular appli
cations can be selected.