Hierarchical classification with a stochastic competitive evolutionary neural tree

The stochastic competitive evolutionary neural tree (SCENT) is a dynamic tree structured network that is able to provide a hierarchical classification of unlabelled data sets. The SCENT is an extension of the competitive evolutionary neural tree (CENT) with the addition of temperature controlled stochastic noise to enable the network to provide solutions independent of initialisation conditions. The main advantage that the SCENT offers over other hierarchical competitive networks is its ability to self determine the number and structure of the competitive nodes in the network without the need for externally set parameters. The network produces stable classificatory structures by halting its growth using locally calculated heuristics. The results of network simulations are presented over Anderson´s IRIS data set