计算机应用 ›› 2013, Vol. 33 ›› Issue (11): 3067-3070.

• 数据库技术 • 上一篇    下一篇

社会网络环境下的协同推荐方法

李慧1,2,胡云2,3,施珺2   

  1. 1. 中国矿业大学 信息与电气工程学院,江苏 徐州 221008;
    2. 淮海工学院 计算机工程学院,江苏 连云港 222001;
    3. 南京大学 计算机科学与技术系,江苏 南京 210000
  • 收稿日期:2013-05-10 修回日期:2013-07-19 出版日期:2013-11-01 发布日期:2013-12-04
  • 通讯作者: 李慧
  • 作者简介:李慧(1979-),女,江苏连云港人,讲师,博士研究生,主要研究方向;Web挖掘、信息检索;胡云(1978-),女,江苏连云港人,副教授,博士研究生,主要研究方向:社会网络分析;施珺(1963-),女,安徽桐城人,副教授,硕士,主要研究方向:智能信息化处理。
  • 基金资助:
    江苏省高校自然科学研究项目;江苏高校优势学科建设项目

Collaborative recommendation algorithm under social network circumstances

LI Hui1,2,HU Yun1,3,SHI Jun1   

  1. 1. School of Computer Engineering, Huaihai Institute of Technology, Lianyungang Jiangsu 222001, China;
    2. School of Information and Electrical Engineering, China University of Mining and Technology, Xuzhou Jiangsu 221008, China;
    3. Department of Computer Science and Technology, Nanjing University, Nanjing Jiangsu 210000, China
  • Received:2013-05-10 Revised:2013-07-19 Online:2013-12-04 Published:2013-11-01
  • Contact: LI Hui

摘要: 针对传统协同过滤推荐算法的数据稀疏性及恶意评分等问题,提出了一种融合信任度与矩阵分解技术实现社会网络推荐的方法。首先通过计算节点的声望值与偏见值发现网络中的不可信节点,并将其评分权重进行弱化。然后将用户-评分矩阵与信任度矩阵相结合,实现社会网络环境下的协同推荐。实验表明,相对于传统的协同过滤算法,该算法可以消减虚假评分或恶意评分给推荐系统带来的负面影响,有效地缓解数据稀疏性与冷启动问题,显著提高推荐系统的推荐质量。

关键词: 社会网络, 权威, 可信度, 矩阵分解, 推荐

Abstract: Concerning data sparsity and malicious behavior of traditional collaborative filtering algorithm, a new social recommendation method combining trust and matrix factorization was proposed in this paper. Firstly, the incredible nodes in the network were founded by computing their prestige value and bias value, and then the weight of their evaluation would be weakened. Finally, the collaborative recommendation was conducted under the social network circumstance by combining the user-item matrix and trust matrix. The experimental results show that the proposed algorithm reduces the importance of not credible node to weaken the negative influence the false or malicious score brings to recommendation system, the data sparsity and malicious behavior problems can be alleviated, and a higher prediction accuracy than that of the traditional collaborative filtering algorithms can be achieved.

Key words: social network, authority, credibility, matrix factorization, recommendation

中图分类号: