Pattern recognition techniques in food quality and authenticity: A guide on how to process multivariate data in food analysis

Abstract

According to the Food and Agriculture Organization of the United Nations, food safety is defined as “assurance that food will not cause harm to the consumer which is prepared and/or eaten according to its intended use”. Regulatory authorities, food producers, and consumers are very interested in food authenticity certification, imposing the need to establish new approaches for identifying and assessing food quality markers. In this review, we present a description of the main analytical techniques and data acquisition used in food analysis, the mostly employed pattern recognition chemometric tools, a survey of applications of different methodologies that have been developed up to date, practical examples, and a critical analysis of advantages and disadvantages of the processing multivariate data in food analysis.