Application of Neural Networks and Meta-Learners to Recognize Beef from OTM Cattle by Using Volatile Organic Compounds

Geographic origin discrimination of pork from different Chinese regions using mineral elements analysis assisted by machine learning techniques

Porkis thelargest-producedandmost-consumedmeat intheworld, and the food market globalization has increased public attention to food origin. Therefore, advanced techniques are required to determine the geographical origin of pork. This study investigated the prospects of using fingerprint analysis of mineral elements and machine learning to facilitate the traceability of pork origin in China. The results showed that each of seven regions had a characteristic element content profile. To improve the performance of the origin traceability model, popular machine learning techniques in food authenticity were introduced. This resulted in a high-performance origin tracing model. Comparing various machine learning algorithms, the feedforward neural network achieved superior performance with an overall accuracy of 95.71% and area under the curve close to one. Thus, this study proves that mineral elements analysis assisted by machine learning can be applied to distinguish pork samples within a country.

Alternative data mining/machine learning methods for the analytical evaluation of food quality and authenticity - A review

In recent years, the variety and volume of data acquired by modern analytical instruments in order to conduct a better authentication of food has dramatically increased. Several pattern recognition tools have been developed to deal with the large volume and complexity of available trial data. The most widely used methods are principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA), soft independent modelling by class analogy (SIMCA), k-nearest neighbours (kNN), parallel factor analysis (PARAFAC), and multivariate curve resolution-alternating least squares (MCR-ALS). Nevertheless, there are alternative data treatment methods, such as support vector machine (SVM), classification and regression tree (CART) and random forest (RF), that show a great potential and more advantages compared to conventional ones. In this paper, we explain the background of these methods and review and discuss the reported studies in which these three methods have been applied in the area of food quality and authenticity. In addition, we clarify the technical terminology used in this particular area of research.