The diversified development of sports clothing styles brings diversified choices to consumers.In order to recommend clothing styles in line with consumers'preferences,the method of category recommendation based on style characteristics is put forward.The Analytic Hierarchy Process is used to determine the weight of attributes,and the key at-tribute factors are ranked to verify the importance of style attributes.Through data collection and modular division,the style parts are disassembled.After encoding the style features,a mathematical model based on vector representation is es-tablished.Combined with K-means clustering and BP neural network for identification and classification,it can effectively reduce the interference of human factors.The cosine similarity is used to calculate the similarity between clothing and cate-gory,and the recommendation is reasonable according to the data characteristics of clothing style and category.Similarity categories facilitate personalized recommendations based on consumer preference for clothing styles.
关键词
运动外套/款式特征/属性划分/类别推荐/BP神经网络
Key words
sport coats/style characteristics/attribute division/category recommendation/BP neural network
维普
万方数据