计算机应用 ›› 2017, Vol. 37 ›› Issue (9): 2531-2535.DOI: 10.11772/j.issn.1001-9081.2017.09.2531

• 先进计算 • 上一篇    下一篇

基于移动用户上下文相似度的张量分解推荐算法

余可钦1,2, 吴映波1,2, 李顺1,2, 蒋佳成1,2, 向德1,2, 王天慧1,2   

  1. 1. 信息物理社会可信服务计算教育部重点实验室(重庆大学), 重庆 400030;
    2. 重庆大学 软件学院, 重庆 401331
  • 收稿日期:2017-01-20 修回日期:2017-03-12 出版日期:2017-09-10 发布日期:2017-09-13
  • 通讯作者: 周伟,wyb@cqu.edu.cn
  • 作者简介:余可钦(1991-),女,湖南益阳人,硕士研究生,CCF会员,主要研究方向:上下文感知推荐系统;吴映波(1978-),男,湖北通城人,副教授,博士,CCF会员,主要研究方向:服务计算、软件服务工程;李顺(1991-),男,河南安阳人,博士研究生,CCF会员,主要研究方向:服务计算、数据挖掘;蒋佳成(1992-),男,四川双流人,硕士研究生,CCF会员,主要研究方向:云计算、虚拟机放置;向德(1992-),男,重庆石柱人,硕士研究生,CCF会员,主要研究方向:大数据流式计算的资源调度;王天慧(1994-),男,湖南邵阳人,硕士研究生,CCF会员,主要研究方向:用户行为分析与推荐系统。
  • 基金资助:
    国家十二五科技支撑计划项目(2014BAH25F01);国家自然科学青年基金项目(71301177);中央高校基本科研业务费资助项目(106112014CDJZR008823);重庆市基础科学与前沿技术研究项目(cstc2013jcyjA1658)。

Tensor factorization recommendation algorithm based on context similarity of mobile user

YU Keqin1,2, WU Yingbo1,2, LI Shun1,2, JIANG Jiacheng1,2, XIANG De1,2, WANG Tianhui1,2   

  1. 1. Key Laboratory of Dependable Service Computing in Cyber Physical Society(Chongqing University), Ministry of Education, Chongqing 400030, China;
    2. School of Software Engineering, Chongqing University, Chongqing 401331, China
  • Received:2017-01-20 Revised:2017-03-12 Online:2017-09-10 Published:2017-09-13
  • Supported by:
    This work is partially supported by the National Science and Technology Support Program of China (2014BAH25F01), the National Natural Science Foundation of China (71301177), the Fundamental Research Funds for the Central Universities (106112014CDJZR008823), the Basic and Advanced Research Program of Chongqing (cstc2013jcyjA1658).

摘要: 针对移动服务推荐中用户上下文环境复杂多变和数据稀疏性问题,提出一种基于移动用户上下文相似度的张量分解推荐算法——UCS-TF。该算法组合用户间的多维上下文相似度和上下文相似可信度,建立用户上下文相似度模型,再对目标用户的K个邻居用户建立移动用户-上下文-移动服务三维张量分解模型,获得目标用户的移动服务预测值,生成移动推荐。实验结果显示,与余弦相似性方法、Pearson相关系数方法和Cosine1改进相似度模型相比,所提UCS-TF算法表现最优时的平均绝对误差(MAE)分别减少了11.1%、10.1%和3.2%;其P@N指标大幅提升,均优于上述方法。另外,对比Cosine1算法、CARS2算法和TF算法,UCS-TF算法在数据稀疏密度为5%、20%、50%、80%上的预测误差最小。实验结果表明UCS-TF算法具有更好的推荐效果,同时将用户上下文相似度与张量分解模型结合,能有效缓解评分稀疏性的影响。

关键词: 用户上下文, 上下文相似度模型, 数据稀疏, 张量分解算法, 移动服务推荐

Abstract: To solve the problem of complex context and data sparsity, a new algorithm for the tensor decomposition based on context similarity of mobile user was proposed, namely UCS-TF (User-Context-Service Tensor Factorization recommendation). Multi-dimensional context similarity model was established with combining the user context similarity and confidence of similarity. Then, K-neighbor information of the target user was applied to the three-dimensional tensor decomposition, composed by user, context and mobile-service. Therefore, the predicted value of the target user was obtained, and the mobile recommendation was generated. Compared with cosine similarity method, Pearson correlation coefficient method and the improved Cosine1 model, the Mean Absolute Error (MAE) of the proposed UCS-TF algorithm was reduced by 11.1%, 10.1% and 3.2% respectively; and the P@N index of it was also significantly improved, which is better than that of the above methods. In addition, compared with Cosine1 algorithm, CARS2 algorithm and TF algorithm, UCS-TF algorithm had the smallest prediction error on 5%, 20%, 50% and 80% of data density. The experimental results indicate that the proposed UCS-TF algorithm has better performance, and the user context similarity combining with the tensor decomposition model can effectively alleviate the impact of score sparsity.

Key words: user context, context similarity model, data sparseness, tensor decomposition algorithm, mobile service recommendation

中图分类号: