基于Vague集理论一维属性——需求匹配的知识推荐算法研究  

Research on Knowledge——Based Recommendation by One-dimensional of Properties and Requirement Matching Based on Vague Sets

作　　者：臧振春 崔春生 ZANG Zhen-chun;CUI Chun-sheng(School of Mathematics and Statistics, Zhoukou Normal University, Zhoukou 466001, China;Department of Computer and Information Engineering, Henan University of Economics and Law, Zhengzhou 450046, China)

机构地区：[1]周口师范学院数学与统计学院,河南周口466001 [2]河南财经政法大学计算机与信息工程学院,河南郑州450046

出　　处：《运筹与管理》2020年第8期112-119,共8页Operations Research and Management Science

基　　金：河南高校哲学社会科学应用研究重大项目计划(2020-YYZD-02);河南省哲学社会科学规划资助项目(2021BJJ010)。

摘　　要：论文从推荐系统中知识推荐算法的核心——产品属性与用户需求的匹配出发,首先探讨了如何采用形式化的Vague集语言描述用户的需求和产品的属性问题。之后分别运用用户兴趣度模型和产品特征模型搭建了用户需求和产品属性的Vague集模型。在模型融合的过程中采用Vague集理论中成熟的相似度计算公式,实现了用户需求与产品属性的匹配计算。最后,从爱奇艺中任意提取了5名注册用户和5部2019年新上映的电影,按照搭建的模型进行了数据计算,得到了可靠的计算结果,同时也构造了用户和产品之间的知识库,为后期知识推荐规则的形成奠定了基础。Beginning with the matching of products properties and user requirements,which is the core of knowledge-based recommendation algorithm in the recommendation system,this paper firstly discusses how to use formal Vague language to describe the user requirements and products properties.Then,a Vague set model of user requirements and products properties are built by user interest model and products characteristic model.In order to get the user requirements and products properties matching calculation,the mature similarity formula in Vague set is used in the process of model fusion.Finally,five registered users and five new movies released in 2019 from iQIYI are extracted to calculate according to the model.The reliable results are got,and at the same time the knowledge base between users and products is made.Therefore,we lay the foundation for the formation of later knowledge recommendation rules.

关 键 词：推荐系统 知识推荐 VAGUE集 产品描述 用户兴趣度 

分 类 号：F713.36[经济管理—产业经济]