Hierarchical classification with a competitive evolutionary neural tree

A new, dynamic, tree structured network, the Competitive Evolutionary Neura l Tree (CENT) is introduced. The network is able to provide a hierarchical classification of unlabelled data sets. The main advantage that the CENT of fers over other hierarchical competitive networks is its ability to self de termine the number, and structure, of the competitive nodes in the network, without the need for externally set parameters. The network produces stabl e classificatory structures by halting its growth using locally calculated heuristics. The results of network simulations are presented over a range o f data sets, including Anderson's IRIS data set. The CENT network demonstra tes its ability to produce a representative hierarchical structure to class ify a broad range of data sets. (C) 1999 Elsevier Science Ltd. All rights r eserved.