基于本体构建的协同推荐研究

Cooperative Recommendation Based on Ontology Construction

通过构建领域本体,利用用户兴趣与领域本体中概念的映射关系,构建用户兴趣本体,发掘用户兴趣模式。研究用户兴趣本体相似度的计算方法,并通过用户兴趣相似度进行垂直加权,通过时间新颖度进行水平加权,从而利用改进的加权关联规则挖掘算法对用户感兴趣的领域本体中的概念进行挖掘,实现面向内容的协同推荐。

This paper constructs domain Ontology firstly, and constructs user interest Ontology by the representation between user interest and domain Ontology, then construct user interest mode. h studies user interest Ontology similarity measure, and identifies the vertical weight by user interest similarity, the horizontal weight by time novelty. At last, it uses the improved weighted association rule algorithm to mine the concepts in domain Ontology, and realizes the cooperative recommendation oriented to content.