APPLICATION AND DEVELOPMENT OF RADIAL BASIS FUNCTION ARTIFICIAL NEURAL NETWORKS TO ANALYSIS AND MODELING OF SPATIALLY DISTRIBUTE INFORMATION (PREPRINT IBRAE-1999-06)

Preprint IBRAE-99-06

Polishchuk V.V., Kanevski M.F. 

Application of radial basis function neural networks to spatailly distributed data is studied. Geostatistics is used to qualify the results. Network extention allows to obtain the estimate variance. «Thick contour» isolines are presented as the outputs for decision making. 

Bibliographical reference

Polishchuk V.V., Kanevski M.F. APPLICATION AND DEVELOPMENT OF RADIAL BASIS FUNCTION ARTIFICIAL NEURAL NETWORKS TO ANALYSIS AND MODELING OF SPATIALLY DISTRIBUTE INFORMATION. Preprint IBRAE-99-06. Moscow: Nuclear Safety Institute. 1999. 27 p. — Refs.: 19 items.