Multichannel Data Aggregation by Layers of Formal Neurons
Principles of separable aggregation of multichannel (multisource) data sets by parallel layers of formal neurons are considered in the paper. Each data set contains such feature vectors which represent objects assigned to one of a few categories.The term multichannel data sets means that each single object is characterised by data obtained through different information channels and represented by feature vectors in a different feature space. Feature vectors from particular feature spaces are transformed by layers of formal neurons what results in the aggregation of some feature vectors. The postulate of separable aggregation is aimed at the minimization of the number of different feature vectors under the condition of preserving the categories separabilty.
KeywordsFeature Vector Feature Space Criterion Function Formal Neuron Feature Subspace
Unable to display preview. Download preview PDF.
- 2.Duda, O.R., Hart, P.E., Stork, D.G.: Pattern Classification. J. Wiley, New York (2000)Google Scholar
- 3.Bobrowski, L.: Piecewise-Linear Classifiers, Formal Neurons and Separability of the Learning Sets. In: Proceedings of ICPR 1996, Wienna, Austria, pp. 224–228 (1996)Google Scholar
- 4.Bobrowski, L.: Eksploracja danych oparta na wypuk’ych i odcinkowo-liniowych funkcjach kryterialnych (Data mining based on convex and piecewise linear (CPL) criterion functions ) (in Polish), Technical University Białystok (2005)Google Scholar