高级检索
    贾冬艳 张付志. 基于双重邻居选取策略的协同过滤推荐算法[J]. 计算机研究与发展, 2013, 50(5): 1076-1084.
    引用本文: 贾冬艳 张付志. 基于双重邻居选取策略的协同过滤推荐算法[J]. 计算机研究与发展, 2013, 50(5): 1076-1084.
    Jia Dongyan and Zhang Fuzhi. A Collaborative Filtering Recommendation Algorithm Based on Double Neighbor Choosing Strategy[J]. Journal of Computer Research and Development, 2013, 50(5): 1076-1084.
    Citation: Jia Dongyan and Zhang Fuzhi. A Collaborative Filtering Recommendation Algorithm Based on Double Neighbor Choosing Strategy[J]. Journal of Computer Research and Development, 2013, 50(5): 1076-1084.

    基于双重邻居选取策略的协同过滤推荐算法

    A Collaborative Filtering Recommendation Algorithm Based on Double Neighbor Choosing Strategy

    • 摘要: 协同过滤是电子商务推荐系统中应用最成功的推荐技术之一,但是传统的协同过滤推荐算法存在推荐精度低和抗攻击能力差的缺陷.针对这些问题,提出了一种基于双重邻居选取策略的协同过滤推荐算法.首先基于用户相似度计算的结果,动态选取目标用户的兴趣相似用户集.然后提出了一种用户信任计算模型,根据用户的评分信息,计算得到目标用户对兴趣相似用户的信任度,并以此作为选取可信邻居用户的依据.最后,利用双重邻居选取策略,完成对目标用户的推荐.实验结果表明该算法不仅提高了系统推荐精度,而且具有较强的抗攻击能力.

       

      Abstract: Collaborative filtering is the most successful and widely used recommendation technology in E-commerce recommender system. It can recommend products for users by collecting the preference information of similar users. However, the traditional collaborative filtering recommendation algorithms have the disadvantages of lower recommendation precision and weaker capability of attack-resistance. In order to solve the problems, a collaborative filtering recommendation algorithm based on double neighbor choosing strategy is proposed. Firstly, on the basis of the computational result of user similarity, the preference similar users of target user are chosen dynamically. Then the trust computing model is designed to measure the trust relation between users according to the ratings of similar users. The trustworthy neighbor set of target user is selected in accordance with the degree of trust between users. Finally, a novel collaborative filtering recommendation algorithm based on the double neighbor choosing strategy is designed to generate recommendation for the target user. Using the MovieLens and Netflix dataset, the performance of the novel algorithm is compared with that of others from both sides of recommendation precision and the capability of attack-resistance. Experimental results show that compared with the existing algorithms, the proposed algorithm not only improves the recommendation precision, but also resists the malicious users effectively.

       

    /

    返回文章
    返回