Design and implementation of popular quality analysis system for dry red wines originating from Northwest China.

This article focuses on the development of a Wine Popular Quality Analysis System (WPQAS) designed to objectively determine the popular quality attributes—such as origin, variety, and vintage year—of dry red wines from Northwest China using physicochemical parameters and machine learning algorithms. Inspired by the French Appellation d'Origine Contrôlée (A.O.C) system, WPQAS employs a Browser/Server three-tier architecture integrating K-nearest neighbour (KNN) for outlier detection, random forest (RF) for parameter weighting, and artificial neural networks (ANN) for classification, achieving an accuracy of approximately 96.7% in identifying wine origin and year. The system includes a scalable database of 600 wine samples from 53 wineries, supports multiple user groups including regulators, wineries, dealers, and consumers, and offers functionalities for model training, prediction, and statistical analysis through an interactive interface. By decoupling models from the core system, WPQAS enhances flexibility, reusability, and ease of application, aiming to address challenges related to wine fraud and to support transparent quality evaluation in the Chinese wine market.