A massively parallel architecture for a self-organizing neural pattern recognition machine
Computer Vision, Graphics, and Image Processing
Introduction to the theory of neural computation
Introduction to the theory of neural computation
Two soft relatives of learning vector quantization
Neural Networks
A Comparative Study of Three Neuronal Networks that use Soft Competition
IWANN '96 Proceedings of the International Workshop on Artificial Neural Networks: From Natural to Artificial Neural Computation
A self-organizing neural tree for large-set pattern classification
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
A new dynamic tree structured network—the StochasticCompetitive Evolutionary Neural Tree (SCENT) is introduced. Thenetwork is able to provide a hierarchical classification ofunlabelled data sets. The main advantage that SCENT offers over otherhierarchical competitive networks is its ability to self-determinethe number and structure of the competitive nodes in the networkwithout the need for externally set parameters. The network producesstable classificatory structures by halting its growth using locallycalculated, stochastically controlled, heuristics. The performance ofthe network is analysed by comparing its results with that of a goodnon-hierarchical clusterer, and with three other hierarchicalclusterers and its non stochastic predecessor. SCENT's classificatorycapabilities are demonstrated by its ability to produce arepresentative hierarchical structure to classify a broad range ofdata sets.