| Abstract
| - The suitability of an approach for extracting heuristic rules from trained artificial neural networks (ANNs)pruned by a regularization method and with architectures designed by evolutionary computation for quantifyinghighly overlapping chromatographic peaks is demonstrated. The ANN input data are estimated by theLevenberg−Marquardt method in the form of a four-parameter Weibull curve associated with the profile ofthe chromatographic band. To test this approach, two N-methylcarbamate pesticides, carbofuran and propoxur,were quantified using a classic peroxyoxalate chemiluminescence reaction as a detection system forchromatographic analysis. Straightforward network topologies (one and two outputs models) allow the analytesto be quantified in concentration ratios ranging from 1:7 to 5:1 with an average standard error of predictionfor the generalization test of 2.7 and 2.3% for carbofuran and propoxur, respectively. The reduced dimensionsof the selected ANN architectures, especially those obtained after using heuristic rules, allowed simplequantification equations to be developed that transform the input variables into output variables. Theseequations can be easily interpreted from a chemical point of view to attain quantitative analytical informationregarding the effect of both analytes on the characteristics of chromatographic bands, namely profile,dispersion, peak height, and residence time.
|
