The work aims to design an incense burner incorporating cultural elements, so as toimprove the design efficiency and accuracy by integrating machine learning algorithm. Firstly, the Likert scale method was used to design the questionnaire and conduct a questionnaire survey. After collection of enough questionnaire feedback, the Particle Swarm Optimization-Support Vector Regression (PSO-SVR) method was used to model the nonlinear combination of product feature-user feedback and obtain the optimal model. Based on the optimized model, the optimal combination of product characteristics was obtained and the product design was carried out. The results show that the user satisfaction of the incense burner design is high and the design process is more convenient. In addition, the user evaluation results of the optimal design closely match the predicted results of the system, which verifies the feasibility and advancement of the proposed method. This study better balances the subjective experience of designers and objective information of users, improves the accuracy of product design decisions, and helps designers develop innovative products that are closely related to user needs.