This paper describes a general fuzzy min-max (GFMM) neural network which is
a generalization and extension of the fuzzy min-max clustering and classif
ication algorithms developed by Simpson, The GFMM method combines the super
vised and unsupervised learning within a single training algorithm. The fus
ion of clustering and classification resulted in an algorithm that can he u
sed as pure clustering, pure classification, or hybrid clustering classific
ation. This hybrid system exhibits an interesting property of finding decis
ion boundaries between classes while clustering patterns that cannot be sai
d to belong to any of existing classes. Similarly to the original algorithm
s, the hyperbox fuzzy sets are used as a representation of clusters and cla
sses. Learning is usually completed in a few passes through the data and co
nsists of placing and adjusting the hyperboxes in the pattern space which i
s referred to as an expansion-contraction process, The classification resul
ts can be crisp or fuzzy. New data can be included without the need for ret
raining. While retaining all the interesting features of the original algor
ithms, a number of modifications to their definition have been made in orde
r to accommodate fuzzy input patterns in the form of lower and upper bounds
, combine the supervised and unsupervised learning, and improve the effecti
veness of operations.
A detailed account of the GFMM neural network, its comparison with the Simp
son's fuzzy min-max neural networks, a set of examples, and an application
to the leakage detection and identification in water distribution systems a
re given.