Independent Component Analysis (ICA) in the deconvolution of overlapping HPLC aromatic peaks of oilIndependent Component Analysis (ICA) in the deconvolution of overlapping HPLC aromatic peaks of oil Πλήρης Δημοσίευση σε Συνέδριο Conference Full Paper 2015-11-012003enChemometric methods like Principal Component Analysis (PCA) and Evolving Factor Analysis (EFA) have been applied to improve peaks separation, especially in HPLC UV-DAD analysis. In this work, the Independent Component Analysis (ICA) was adopted for the separation of overlapping aromatic peaks and the simultaneous determination of the underlying spectra. The application of the method on middle petroleum fractions showed an improved separation between the aromatic groups compared to the previously used methods. The determined by the ICA method UV spectra of the component groups can be further exploited for the compositional characterization of the aromatic fractions.http://creativecommons.org/licenses/by/4.0/96-993rd International Conference, Instrumental Methods of Analysis. Modern Trends and ApplicationsPasadakis_et_al_IMA_2003.pdfChania [Greece]Library of TUC2015-11-01application/pdf399.8 kBfree Pasadakis Nikos Πασαδακης Νικος Gaganis Vasileios Γαγανης Βασιλειος Smaragdis, Paris Mitsubishi Electric Research Laboratories ICA Independent Component Analysis