Personalized Recommendation Algorithm for Cultural and Creative Products Based on Fuzzy Decision Support System

Cultural and creative products are extensive, difficult to mass produce, and undersupplied. They can demonstrate their distinctive qualities through multifunctional design while raising cultural and economic value. Three circumstances often emerge when cultural and creative customers purchase stationery: purchasing inclination, purchasing behavior, and reflection. Limited mass manufacturing, undersupply, and the need to balance cultural and commercial value are common cultural and creative product issues. Hence, this study proposes a fuzzy context-aware neural recommendation algorithm (F-CANRA) to analyze consumers’ purchase behaviors regarding cultural and creative products (CCPs). A graph neural network (GNN) is established by studying users’ cultural and creative product-buying behavior. In this environment, a fuzzy decision support system (FDSS) could assist in making product recommendations for artistic and innovative actions (e.g., artwork, music, crafts, etc.) by taking into account a variety of criteria, including users’ preferences, current trends, cultural importance, and the fuzziness of these factors (e.g., a user’s approximate preference for “modern yet traditional” designs). This article proposes a decision-making framework utilizing a fuzzy decision support system (FDSS) integrated with a fuzzy analytic hierarchy process (FAHP). This framework aims to prioritize design elements, develop cultural and creative design components, and identify and analyze the key criteria influencing user needs. This research shows that the strategy may assist industrial designers in creating better color schemes for creative and cultural products by incorporating group users’ visual preferences into purchasing intention via multiuser decision consistency. The simulation outcome demonstrates that the suggested model increases the purchase intention prediction ratio of 97.8%, customer emotional satisfaction ratio of 98.5%, product development ratio of 96.2%, recommendation accuracy ratio of 95.2%, and product design costs of 7.3% compared to other existing models. As these outcomes show, the approach can help companies and industrial designers create CCPs that are culturally important and financially feasible. The F-CANRA platform provides a strong answer to the challenges of tailored product suggestions and well-informed design choices by connecting sophisticated algorithms with the cultural industry's complex requirements.