ChemNet
 
Previous article Next article Contents  

Ya. N. Pushkarova, A. B. Sledzevskaya, A. V. Panteleimonov, N. P. Titova, O. I. Yurchenko, V. V. Ivanov, Yu. V. Kholin

Identification of water samples from different SPRINGS and rivers of Kharkiv: comparison of methods of multivariate data analysis

Abstract

Application of artificial neural networks for identification of water samples from different Kharkiv springs and rivers (Eastern Ukraine) was studied. The initial experimental data consisted of metal ions concentrations in the water samples. The artificial neural networks algorithms were demonstrated to provide the correct identification of water samples even in the presence of gaps in the initial data. To realize successfully each particular neural network it is necessary to determine the optimal number of neurons. The corresponding rule was proposed.
Key words: qualitative chemical analysis, identification, artificial neural network, linear discriminant analysis.
Moscow University Chemistry Bulletin.
2012, Vol. 53, No. 6, P. 405
   

Copyright (C) Chemistry Dept., Moscow State University, 2002
   Overview
   Editorial board
   Tables of Contents
   Subscription

The site is supported by Russian Foundation for Basic Research
  The using of published on this page materials is not allowed without special permission
Copyright (C) Chemisty Department of Moscow State University
Web-Editor: B.I.Pokrovskii
Web-design: Copyright (C) MIG and VVM
webmaster@www.chem.msu.su