RT Journal Article
T1 Effectiveness of machine learning algorithms as a tool to meat traceability system. A case study to classify Spanish Mediterranean lamb carcasses
A1 García Infante, Manuel
A1 Castro Valdecantos, Pedro
A1 Delgado Pertiñez, Manuel
A1 Teixeira, Alfredo
A1 Guzmán Guerrero, José Luis
A1 Horcada, Alberto
AB Establishing the traceability of meat products has been a major focus of food science in recent decades. In thiscontext, recent advances in food nutritional biomarker identification and improvements in statistical technologyhave allowed for more accurate identification and classification of food products. Moreover, artificial intelligencehas now provided a new opportunity for optimizing existing methods to identify animal products. This studypresents a comparative analysis of the effectiveness of different machine learning algorithms based on raw datafrom analyses of organoleptic, sensory and nutritional meat traits to differentiate categories of commercial lambfrom an indigenous Spanish breed (Mallorquina breed) obtained from the following production systems: sucklinglambs; light lambs from grazing; and light lambs from grazing supplemented with grain. Six machine learningalgorithms were evaluated: Artificial Neural Network (ANN), Decision Tree, K-Nearest Neighbours (KNN), NaiveBayes, Multinomial Logistic Regression, and Support Vector Machine (SVM). For each algorithm, we tested threedatasets, namely organoleptic traits and sensorial traits (CIELAB colour, water holding capacity, Warner-Bratzlershear force, volatile compounds and trained tasters), and nutritional traits (proximate composition and fatty acidprofile). We also tested a combination of all three datasets. All the data were combined into a dataset with 144variables resulting from the meat characterization, which included 11,232 event records. The ANN algorithmstood out for its high score with each of the three datasets used. In fact, we obtained an overall accuracy of 0.88,0.83, and 0.88 for the organoleptic-sensory, nutritional, and combined datasets, respectively. The effectiveness ofusing the SVM algorithm to assign categories of lambs according to its production system performed better withnutritional traits and the full characterization, with performances equal to those obtained with ANN. The KNNalgorithm showed the worst performance, with overall accuracies of 0.54 or lower for each of the datasets used.The results of this study demonstrate that machine learning is a useful tool for classifying commercial lambcarcasses. In fact, the ANN and SVM algorithms could be proposed as tools for differentiating categories of lambproduction based on the organoleptic, sensory and nutritional characteristics of Mediterranean light lambs’ meat.However, in order to improve the traceability methods of lamb meat production systems as a guarantee forconsumers and to improve the learning processes used by these algorithms, more studies along these lines withother lamb breeds are required
PB Elsevier
SN 0956-7135
SN 1873-7129 (electrónico)
YR 2024
FD 2024-06
LK https://hdl.handle.net/10272/24319
UL https://hdl.handle.net/10272/24319
LA eng
NO García-Infante, M., Castro-Valdecantos, P., Delgado-Pertíñez, M., Teixeira, A., Guzmán, J. L., & Horcada, A. (2024). Effectiveness of machine learning algorithms as a tool to meat traceability system. A case study to classify Spanish Mediterranean lamb carcasses. In Food Control (Vol. 164, p. 110604). Elsevier BV. https://doi.org/10.1016/j.foodcont.2024.110604
NO This research has been financed by the Institute for Agricultural andFisheries Research and Training (IRFAP) of the Government of theBalearic Islands (PRJ201502671-0781), the Spanish National Instituteof Agricultural and Food Research and Technology and the EuropeanSocial Fund (FPI2014-00013). Particular gratefulness to PhD OlivaPolvillo Polo (CITIUS, University of Seville’s Centre for Research) forcontributing her knowledge in chromatography analysis
DS Repositorio Institucional de la Universidad de Huelva
RD 30 may 2026