Sigma-if neural network as the use of selective attention technique in classification and knowledge discovery problems solving

Abstract

The article presents the most important properties of Sigma-if neuron and neural network, which use a selective attention technique to solve classification problems. Abilities of Sigma-if neuron to perform active aggregation of input signals and to solve linearly inseparable problems are discussed. Variety of conducted experiments, during which Sigma-if network was compared with multilayer perceptron, are also presented. These experiments show benefits from using Sigma-if network instead of MLP, both in classification problems solving and in knowledge discovery from data.