Abstract This paper mainly utilizes a knowledge graph to construct a two-way fusion clothing design system. The article is based on the model to extract the knowledge graph relationship of clothing and apparel, and the apriori association rule algorithm is used to mine the user’s preferences, and then the clothing is designed by digital technology. In the end, it examines the interconnectedness of the knowledge graph and the effect of recognition on different fabrics. The consumer’s satisfaction and attention to the impact of digitalized clothing design were studied, and the accuracy of clothing design size was explored. The study showed that the clothing design size error was kept within ±2%, and the overall customer satisfaction with the clothing design was between (3.9 and 4.35) with high overall satisfaction. This study can improve the design quality of apparel brands, reduce production costs, and provide new ideas and directions for future apparel design in the digital era.