Decision boundary modelling-a constructive algorithm for high precision real world data models

A network´s generalisation is extrapolated from the training examples, i.e. it is defined by the network´s model of the training data. Simplifying this model will increase the resistance to noise, inherent in the training data, but will also lose information which may not be noise. This paper discusses the use of Voronoi tessellations for highly detailed data models and the effects of noise. An original algorithm is developed for constructing a competitive layer of k-nearest neighbour type neurons. These neurons are trained to model the decision boundaries and therefore approximate the perfect bisector that is required. The new algorithm is empirically compared with three existing hyperplanic algorithms-neural tree network (NTN), backpropagation (BP), and the entropy net, on three real world continuous data sets