计算机应用 ›› 2015, Vol. 35 ›› Issue (7): 1979-1983.DOI: 10.11772/j.issn.1001-9081.2015.07.1979

• 人工智能 • 上一篇    下一篇

融合朋友关系和标签信息的张量分解推荐算法

丁小焕, 彭甫镕, 王琼, 陆建峰   

  1. 南京理工大学 计算机科学与工程学院, 南京 210094
  • 收稿日期:2015-01-20 修回日期:2015-03-26 出版日期:2015-07-10 发布日期:2015-07-17
  • 通讯作者: 丁小焕(1991-),女,浙江衢州人,硕士研究生,主要研究方向:数据挖掘,xiaohuan849@sina.com
  • 作者简介:彭甫镕(1987-),男,贵州遵义人,博士研究生,主要研究方向:数据挖掘、推荐系统; 王琼(1981-),女,江苏南京人,副教授,博士,主要研究方向:模式识别、智能系统; 陆建峰(1969-),男,江苏淮安人,教授,博士,主要研究方向:智能系统、数据挖掘。
  • 基金资助:

    江苏省"六大人才高峰"高层次人才项目;江苏省研究生科研创新计划项目(KYLX0382)。

Tensor factorization recommendation algorithm combined with social network and tag information

DING Xiaohuan, PENG Furong, WANG Qiong, LU Jianfeng   

  1. School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing Jiangsu 210094, China
  • Received:2015-01-20 Revised:2015-03-26 Online:2015-07-10 Published:2015-07-17

摘要:

针对大众标注网站项目推荐系统中存在数据矩阵稀疏性影响推荐效果的问题,考虑矩阵奇异值分解(SVD)能有效地平滑数据矩阵中的数据,以及朋友圈能够反映出一个人的兴趣爱好,提出了一种融合朋友关系和标签信息的张量分解推荐算法。首先,利用高阶奇异值分解(HOSVD)方法对用户-项目-标签三元组信息进行潜在语义分析和多路降维,分析用户、项目、标签三者间关系;然后,再结合用户朋友关系、朋友间相似度,修正张量分解结果,建立三阶张量模型,从而实现推荐。该模型方法在两个真实数据集上进行了实验,结果表明,所提算法与高阶奇异值分解的方法比较,在推荐的召回率和精确度指标上分别提高了2.5%和4%,因此,所提算法进一步验证了结合朋友关系能够提高推荐的准确率,并扩展了张量分解模型,实现用户个性化推荐。

关键词: 张量分解, 高阶奇异值分解, 朋友关系, 标签, 推荐

Abstract:

The item recommendation precision of social tagging recommendation system was affected by sparse data matrix. A tensor factorization recommendation algorithm combined with social network and tag information was proposed, in consideration of that Singular Value Decomposition (SVD) had good processing properties to deal with sparse matrix, and that friends' information could reflect personal interests and hobbies. Firstly, Higher-Order Singular Value Decomposition (HOSVD) was used for latent semantic analysis and multi-dimensional reduction. The user-project-tag triple information could be analyzed by HOSVD, to get the relationships among them. Then, by combining the relationship of users and friends with the similarity between friends, the result of tensor factorization was modified and the third-order tensor model was set up to realize the item recommendation. Finally, the experiment was conducted on two real data sets. The experimental results show that the proposed algorithm can improve respectively recall and precision by 2.5% and 4%, compared with the HOSVD method. Therefore, it is further verified that the algorithm combining with the relation of friends can enhance the accuracy of recommendation. What's more, the tensor decomposition model is expanded to realize the user personalized recommendation.

Key words: tensor factorization, Higher-Order Singular Value Decomposition (HOSVD), social network, tag, recommendation

中图分类号: