服装推荐系统研究进展

Research Progress of Clothing Recommendation System

利用文献调研和VOSviewer对国内外服装推荐系统相关文献的发文量、关键词进行可视化分析,梳理总结服装推荐系统的发展趋势。在此基础上,从感官评价、模糊技术、协同过滤、机器学习4个方面归纳整理了服装推荐系统关键技术与方法。从应用情况的角度探讨了服装推荐系统的分类,即服装产品推荐、服装搭配推荐与服装设计推荐。最后,基于服装推荐系统发展趋势、关键方法与技术和应用领域,指出未来可在传统文化元素的创新应用、个人服饰搭配与依据消费者隐性需求和显性需求的轻定制交互化设计中深入研究服装推荐系统。

Through literature research and VOSviewer,this paper visually analyzed the volume and keywords of the articles and summarized the development trend of the clothing recommendation systems.On this basis,the key technologies and methods of clothing recommendation systems were summarized from four aspects,containing sensory evaluation,fuzzy technology,collaborative filtering,and machine learning.Then the classification of clothing recommendation systems was discussed from the perspective of the application,namely,clothing product recommendation,clothing matching recommendation,and personalized customization.Based on the system development trend,key methods and technologies,and application fields,it is pointed out that the clothing recommendation system can be further studied in the innovative application of traditional cultural elements,personal clothing collocation and interactive design of light customization according to the implicit and explicit needs of consumers.