Research on the Interactive Effects of Product Recommendation Approaches and Product Types on Consumers’ Purchase Intention–Regulation of Product Involvement

The main purpose is to analyze how different product recommendation approaches and product types influence consumers’ purchasing decisions, elucidate the interactive effects between these variables, and develop platform-specific operational strategies to optimize recommendation systems. Based on the Stimulus-Organism-Response (SOR) theory and the Elaboration Likelihood Model (ELM), this study employs a 2 (product recommendation type: AI vs. human) × 2 (product type: functional vs. hedonic) between-subjects experimental design to examine how recommendation systems interact with product characteristics to influence consumers’ purchase intentions. Specifically, it examines the mediating role of perceived value and the moderating effect of product involvement. An inter-group experimental design was employed to gather data via online questionnaires. ANOVA and other statistical methods were used to analyze the interaction, mediation, and moderation effects among the variables. The findings indicate a significant interaction between recommendation mode and product type. AI recommendations are more effective for functional products, whereas human recommendations are more advantageous for hedonic products. Perceived value mediates this interaction, with quality and price value being predominant in the AI recommendation pathway, while emotional, quality, and price value drive the human recommendation pathway. Moreover, the interaction effect is moderated by product involvement, with the matching effect being significantly enhanced under high involvement conditions, while the difference diminishes under low involvement conditions. This research offers new insights into consumer behavior theory and provides a practical foundation for optimizing recommendation strategies on e-commerce platforms.