Integration of spectroscopy and chemometric analysis for food authentication: A review

Abstract

The high demand in the food sector necessitates an authentication process to verify products and prevent food fraud. In the food industry, applying spectroscopy techniques combined with chemometric analysis has become an efficient tool for authentication, thanks to its rapid results, cost-effectiveness, and reliability. Among the spectroscopy techniques reviewed in this paper are near-infrared (NIR), Fourier transform infrared (FTIR), ultraviolet (UV), and Raman spectroscopy. This paper reviews recent spectroscopy applications to food, such as oils, agricultural products, and beverages. This paper reviews various types of classification and regression algorithms in chemometric analysis to demonstrate the capabilities of these algorithms for use in food authentication processes. The chemometric model can accurately identify authentic samples with minimal risk of misclassification. This review presents a comparative synthesis of algorithms for various types of food samples, highlighting the performance of predictive algorithms. This review will provide a practical framework for researchers seeking to develop a robust predictive system for food authentication using spectroscopy and chemometrics. The review discusses the emerging trend of integrating spectroscopy into real-time authentication systems, particularly in the application of data fusion and deep learning techniques.