20.500.12556/RUL-180964 Discriminating cinnamon species and geographic origin using HPLC-UV and chemometrics The aim of this study was to evaluate biomarkers for the differentiation and classification of cinnamon species and geographical origin. A reverse phase high performance liquid chromatography method with UV-Vis detection was used to quantify the following biomarkers: coumarin, cinnamic acid, cinnamaldehyde and eugenol. A total of 23 cinnamon samples were analysed and processed using chemometric methods. Besides commonly used principal component analysis, uniform manifold approximation and projection was also used for exploratory analysis. Next, combination of dimensions reduction and classification algorithms were used for the development of 16 classification models. The models were validated using the bootstrapping resampling method. The best classification was achieved with the combination of principal component analysis - regularised discriminant analysis model with an overall accuracy of 84.0% and 89.2% in discriminating cinnamon species and their geographical origin, respectively. In addition, the samples were successfully classified according to their shape and whether the cinnamon samples were certified as “bio” organic. Namen te študije je bil ovrednotiti biomarje za razlikovanje in razvrščanje vrst cimeta ter njegovega geografskega porekla. Za kvantifikacijo biomarjev – kumarina, cimetne kisline, cimetnega aldehida in evgenola – je bila uporabljena reverzno-fazna metoda tekočinske kromatografije visoke ločljivosti z UV-Vis detekcijo. Analiziranih je bilo skupno 23 vzorcev cimeta, ki so bili obdelani s kemometričnimi metodami. Poleg splošno uporabljene metode glavnih komponent (PCA) je bil za raziskovalno analizo uporabljen tudi UMAP (angl. uniform manifold approximation and projection). V nadaljevanju je bila uporabljena kombinacija algoritmov za zmanjševanje dimenzionalnosti in klasifikacijo za razvoj 16 klasifikacijskih modelov. Modeli so bili validirani z metodo ponovnega vzorčenja z zaganjanjem (angl. bootstrapping). Najboljša klasifikacija je bila dosežena s kombinacijo modela analize glavnih komponent in regularizirane diskriminantne analize, ki je dosegla skupno natančnost 84,0 % pri razlikovanju vrst cimeta in 89,2 % pri določanju njihovega geografskega porekla. Poleg tega so bili vzorci uspešno razvrščeni glede na njihovo obliko in glede na to, ali so bili certificirani kot ekološki (»bio«) izdelki. cinnamon species liquid chromatography biomarkers chemometrics classification geographical origin vrste cimeta tekočinska kromatografija biomarkerji kemometrija klasifikacija geografsko poreklo true false true Angleški jezik Slovenski jezik Članek v reviji 2026-03-20 13:31:21 2026-03-20 13:31:21 2026-03-23 03:53:33 0000-00-00 00:00:00 2026 0 0 Str. 279-287 no. 1 Vol. 73 2026 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 543.544.5:582.673.42 1580-3155 10.17344/acsi.2025.9540 272403971 RAZ_Markelj_Jernej_2026.pdf RAZ_Markelj_Jernej_2026.pdf 1 68500688EDC827D53201DB02E54A4832 b16a7e2f6e0f8d87246b41083c610c4a166e9e8e0b7c5403785bfb55fccf5166 549f8ed1-2459-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=231001 https://acsi-journal.eu/index.php/ACSi/article/view/9540 1 9dd0c9d7-2458-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=231000 Fakulteta za kemijo in kemijsko tehnologijo 0 0 0