基于个性化推荐的青年女裤知识图谱构建  

Construction of a Knowledge Graph for Young Women'sTrousers Based on Personalized Recommendation

作　　者：刘丁菲 李艳梅[1] LIU Dingfei;LI Yanmei(College of Shanghai University of Engineering and Technology,Shanghai 201620,China)

机构地区：[1]上海工程技术大学纺织服装学院,上海201620

出　　处：《北京服装学院学报（自然科学版）》2024年第3期70-77,共8页Journal of Beijing Institute of Fashion Technology：Natural Science Edition

基　　金：上海艺术科学规划重大项目(2017wNO.107)。

摘　　要：基于个性化推荐需要,对青年女裤知识和实体属性关系进行抽取,从基础、表现和外在3个维度将女裤属性分为13类,组成服装、属性、属性值三元组,利用Neo4j图数据库存储。对已获取到的300组青年女裤数据进行聚类分析,得到15组女裤数据,选取每组中的代表性女裤构建可视化知识图谱,并将其用于个性化女裤推荐,最后通过调研评分验证了该图谱的合理性。Based on the need of personalized recommendations,the knowledge of young women's trousers and the relationship between entity attributes are extracted.The attributes of women's trousers are classified into 13 categories from three dimensions:basic,performance and external,forming a ternary group of garments,attributes,and attribute values,and stored in the Neo4j graph database.The 300 groups of young women's trousers data that have been acquired are clustered and analyzed to obtain 15 groups of women's trousers data,and the representative women's trousers in each group are selected to construct a visual knowledge graph,which is used for personalized women's trousers recommendation,and finally the rationality of the graph is verified by the research scores.

关 键 词：知识图谱 服装推荐 青年女裤 Neo4j 

分 类 号：TS941.26[轻工技术与工程—服装设计与工程]