.: Marcos Gestal :.     -----


Datos de la Publicación:

Marcos Gestal, Mari Paz Gómez, José Manuel Andrade, Julián Dorado, Esther Fernández, Darío Prada, Alejandro Pazos
Título: Selection of variables by Genetic Algorithms and Procrustes Rotation for Classifying Apple Beverages by FTIR and ANNs
Congreso: Conferentia Chemometrica 2003
Lugar Celebración: Budapest (Hungría)
Fecha Celebración: 27-29 de Octubre de 2003
Publicación: Libro de Actas
ISBN: 963-508-391-2
Páginas: 40
Fecha Publicación: Octubre 2003


The importance of juice beverages, and of apple juice in particular, in daily food habits makes juice authentication an important to avoid fraudulent practices and to protect human health. Among the instrumental techniques available in analytical laboratories, MIR spectroscopy is a fast and convenient technique to perform screening studies in order to assess the quantity of pure juice in commercial beverages. IR quantitation of chemicals in complex samples is an ongoing topic in Analytical Chemistry since the use of advanced statistical and computational methods can satisfactorily handle such real problems. Once the IR spectra are digitized the chemical information is ?translated? into a large set of variables which can be studied by powerful mathematical and computational methods in order to get additional valuable information. Although the digitized spectra are inherently useful the number of variables is generally large. Here, genetic algorithms and Procrustes rotation were applied to select a minimum set of wavenumbers to perform classification and regression. Once the variables were selected, they were employed to classify apple beverages using different multivariate methods (PLS, SIMCA, Potential curves, and artificial neural networks).