This article investigates the behaviour of a self-organizing logic neu
ral network when it is tasked with clustering complex data spaces. The
network is based on the discriminator-node structure and is trained u
sing an unsupervised-learning adaptation rule. The network performance
is evaluated by applying it to clustering tasks involving identifiabl
e classes, each of which consists of a large number of distinct subcla
sses. The results presented are supported by a statistical analysis, w
hich indicates that the system is indeed suited to clustering such com
plex data sets.