Wine authenticity verification as a forensic problem: An application of likelihood ratio test to label verification

• Published in: Food chemistry Vol. 150; pp. 287 - 295
• Main Authors: Martyna, Agnieszka, Zadora, Grzegorz, Stanimirova, Ivana, Ramos, Daniel
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.05.2014; Elsevier
• Subjects: Alcoholism and acute alcohol poisoning; analysis; Biological and medical sciences; Chemical composition; classification; Classification problem; Discriminant Analysis; Empirical cross entropy; Entropy; Evaluation of forensic evidence; Food Contamination; Food Contamination - analysis; Food Labeling; Food products authenticity; Foods; Forensic engineering; Likelihood Functions; Likelihood ratio; Medical sciences; Origins; Performance evaluation; product authenticity; statistics & numerical data; Toxicology; Vitaceae; Wine; Wine - analysis; Wine - classification; Wine - statistics & numerical data; Wines; Food products authenticity; Likelihood ratio; Classification problem; Empirical cross entropy; Evaluation of forensic evidence; Wine; Evaluation; Forensic aspect; Verification; Alcoholic beverage; Classification; Ratio; Food
• ISSN: 0308-8146; 1873-7072; 1873-7072
• DOI: 10.1016/j.foodchem.2013.10.111

•27 Parameters describing wine samples were used for verifying their authenticity.•Likelihood ratio approach was used to evaluate their evidential value.•Uni- and multivariate LR models were proposed for solving the classification problem.•Empirical cross entropy was used to measure the performance of LR approach. The aim of the study was to investigate the applicability of the likelihood ratio (LR) approach for verifying the authenticity of 178 samples of 3 Italian wine brands: Barolo, Barbera, and Grignolino described by 27 parameters describing their chemical compositions. Since the problem of products authenticity may be of forensic interest, the likelihood ratio approach, expressing the role of the forensic expert, was proposed for determining the true origin of wines. It allows us to analyse the evidence in the context of two hypotheses, that the object belongs to one or another wine brand. Various LR models were the subject of the research and their accuracy was evaluated by the Empirical cross entropy (ECE) approach. The rates of correct classifications for the proposed models were higher than 90% and their performance evaluated by ECE was satisfactory.