结合资源语义和用户访问路径分析的个性化推荐模型

Personalized Recommendation Model Based on the Combination of Resource Semantic and User Access Path Analysis

针对已有的个性化推荐方法仅根据用户的访问行为,或者仅根据用户所访问的资源主题的相似性进行用户模式划分的问题,提出一种结合资源语义和用户访问路径分析的个性化推荐方法。在对用户访问路径进行语义标注的基础上,同时基于用户访问路径相似度和用户访问主题相似度对用户进行模式划分,并采用混合推荐技术构建个性化推荐模型。初步实验结果表明了方法的有效性和可行性。

In the view of the existing personalized recommendation method, in which divides the user model only according to the user access behavior or based on the similarity of user access topics. This paper puts forward a personalized recommendation method based on the combination of resource semantic and user access path analysis. The paper divides the user model on the basis of the semantic label of user access path, as well as the similarity of user access path and similarity of user access topics. Then, the paper constructs the personalized recommendation model by adopting blending recommendation technology. The preliminary experiment results show the effectiveness and feasibility of the proposed method.