Métodos quimiométricos aplicados em análises de frutas tropicais e derivados: Um estudo de revisão

Abstract

Considering the worldwide increase in the production and consumption of fruits and derivatives, there is a naturally greater demand for product quality, either by the consumer or by the rigidity of the control bodies, which increasingly requires a safer product, free from possible adulteration or fraud. Therefore, the search for methods that help in the required analyses, to guarantee the quality of the product, is expanded and enhanced. In this context, it is common to increase the number of articles published, aspiring to solve possible problems that interfere in the production chain, given the complexity of the studied matrix and the range of interference that affects it. It is known that with the advent of computers and new software, chemometrics emerges as an excellent set of techniques for analyzing data from complex matrices, enabling a better inquiry into sample data. This article carried out a literature review regarding the application of chemometric methods used in food analysis, with specificity in tropical fruits and derivatives, analyzing scientometric indicators of the applications of the main analysis techniques in matrices, as well as bibliometric indicators of the application of chemometric classification and calibration methods. The journal portal of the Coordination for the Improvement of Higher Education Personnel (CAPES) was used for date collection and the Web of Science database was the chosen base for research. We adopted methodologies inherent to bibliometrics that made it possible to gather a larger set of articles and analyze them quantitatively, considering the theme. Regarding the review, we can see that Brazil is one of the countries with the highest number of publications in the sector, followed by China and India. We also found that, among the sets of articles analyzed in the period, Principal Component Analysis (PCA), Partial Least Square Regression (PLS) and Cluster Analysis by Hierarchical Methods (HCA) are the most used methods in data processing, providing a result with greater strength and reliability.