Enhancing wine selection experience for young Chinese consumers: an optimized collaborative filtering recommendation model with integrated consumer profiling

Purpose This article aims to propose an optimized collaborative filtering recommendation model with integrated multi-dimensional consumer profiling, so as to address the limitations of existing personalized wine recommendation methods and improve the wine selection experience for young Chinese consumers. Design/methodology/approach This study adopts the adaptive density peak and label propagation layer-by-layer (ADPLP) clustering algorithm to segment young Chinese wine consumers and construct multi-dimensional group-specific profiles. Then, it proposes an optimized collaborative filtering recommendation model with integrated consumer profiling based on the profile discriminant coefficient, consistency factors of consumer characteristics, and preference complexity, which could be applied to solving personalized wine recommendation issues for young Chinese consumers. Findings The proposed optimized collaborative filtering recommendation model with integrated consumer profiling achieves well performance in precision and novelty when delivering personalized wine recommendations to young Chinese consumers. Originality/value The article overcomes the limitations of existing wine recommendation methods for young Chinese consumers. By incorporating both consumers’ characteristics and the complexity of their wine preferences, this study proposes an optimized collaborative filtering recommendation model with integrated consumer profiling to enhance the wine selection experience of young Chinese consumers by using personalized recommendation method.