Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/121086
Title: Multivariate analysis of mineral profile in paprika with protected designation of origin
Author: López Pascual, Óscar
Tutor: Perez-Alvarez, Nuria  
Others: Ventura, Carles  
Abstract: In an increasingly globalized world, food fraud has brought a growing concern in food producers, distributors and also consumers. In this context, food control laboratories play a key role in fraud detection. One possible type of fraud is related to incorrect labelling of products with a Protected Designation of Origin speci cation. From the analytical point of view, one of the main modern techniques for origin detection is the untargeted multivariate analysis, like mineral pro le, followed by an appropriate statistical treatment of the produced data, which its aim is drawing conclusions regarding the origin of the product. In this work machine learning models have been applied to paprika's mineral pro le data with a protected designation of origin for the obtention of classi catory results in terms of the origin of the products. The applied techniques have been Principal Components Analysis, Cluster Analysis, Discriminant Linear Analysis and, for the fi rst time in this type of product, Random Forest method, which has been able to correctly classify all the analyzed samples according to the geographical origin. A global method has been developed including the acquisition of the mineral pro file by means of Mass Spectrometry, and the data processing algorithm, in R language, which can be applied to other paprika designations and presumably to other types of products with Protected Designation of Origin.
Keywords: food fraud
icp-ms
random forest
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jun-2020
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
olopezpasTFM0620memory.pdfMemory of TFM2,56 MBAdobe PDFThumbnail
View/Open