The application of multi-way models in food analysis as in many other research fields is rapidly increasing, above all because it can significantly help to improve visualization and interpretation of complex multivariate data characterized by different sources of variability. Furthermore, in food authentication tasks a fingerprinting strategy is frequently applied which requires the direct analysis of complex instrumental data, often obtained by hyphenated analytical techniques, which lead to multi-way data. In this work, the main features and advantages of multi-way regression are presented through the study of the evolution of the sensory and compositional profile during the ageing process of Aceto Balsamico Tradizionale di Modena (ABTM), a typical Italian food product, which represents a very interesting benchmark for testing new analytical methodologies due to its long ageing process and the peculiarity of the traditional making procedure. A series of 6 casks for each of 6 different producers has been characterized through sensory and instrumental analysis of the volatile fraction by head space-solid phase micro extraction/gas chromatography (HS-SPME/GC). The N-PLS method has been used as regression method since many sources of variability have to be taken into account, e.g. different cask series, different samples ageing, different panelists, etc. The validity of this choice is evaluated by comparing with results obtained from unfold-PLS method. Satisfactory regression models were obtained, highlighting the efficiency of the three-way model, which was shown to be more robust and interpretable.
Application of N-PLS to gas chromatographic and sensory data of traditional balsamic vinegars of modena / Durante, Caterina; Cocchi, Marina; M., Grandi; Marchetti, Andrea; R., Bro. - In: CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS. - ISSN 0169-7439. - STAMPA. - 83:1(2006), pp. 54-65. [10.1016/j.chemolab.2006.01.004]
Application of N-PLS to gas chromatographic and sensory data of traditional balsamic vinegars of modena
DURANTE, Caterina;COCCHI, Marina;MARCHETTI, Andrea;
2006
Abstract
The application of multi-way models in food analysis as in many other research fields is rapidly increasing, above all because it can significantly help to improve visualization and interpretation of complex multivariate data characterized by different sources of variability. Furthermore, in food authentication tasks a fingerprinting strategy is frequently applied which requires the direct analysis of complex instrumental data, often obtained by hyphenated analytical techniques, which lead to multi-way data. In this work, the main features and advantages of multi-way regression are presented through the study of the evolution of the sensory and compositional profile during the ageing process of Aceto Balsamico Tradizionale di Modena (ABTM), a typical Italian food product, which represents a very interesting benchmark for testing new analytical methodologies due to its long ageing process and the peculiarity of the traditional making procedure. A series of 6 casks for each of 6 different producers has been characterized through sensory and instrumental analysis of the volatile fraction by head space-solid phase micro extraction/gas chromatography (HS-SPME/GC). The N-PLS method has been used as regression method since many sources of variability have to be taken into account, e.g. different cask series, different samples ageing, different panelists, etc. The validity of this choice is evaluated by comparing with results obtained from unfold-PLS method. Satisfactory regression models were obtained, highlighting the efficiency of the three-way model, which was shown to be more robust and interpretable.File | Dimensione | Formato | |
---|---|---|---|
33_Chemolab2006.pdf
Accesso riservato
Tipologia:
AAM - Versione dell'autore revisionata e accettata per la pubblicazione
Dimensione
322.14 kB
Formato
Adobe PDF
|
322.14 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris