协同过滤能够满足用户的偏好,为用户提供个性化的指导,是当前互联网推荐引擎中的核心技术。然而,该技术的发展面临着严重的用户评分稀疏性问题。用户评分历史中包含着丰富的上下文信息,因此该文通过利用两种上下文信息对评分稀疏性问题进行了有益的探索: 利用物品之间的层次关联关系挖掘用户的潜在喜好;对用户评分的短期时间段效应进行建模。并提出了基于两种上下文信息的统一模型CICF。通过在Yahoo音乐数据集上的实验表明,CICF相比传统协同过滤算法能够显著提高预测效果;并通过在不同稀疏度的训练集上的实验证实了CICF能够有效地缓解评分稀疏性问题。
Abstract
Collaborative Filtering (CF) could satisfy users preferences and provide personalized guidance. As the key techniques in current Internet recommendation engines, however, this technology suffers from severe sparse users ratings problem. Considering the plenty context information in users rating histories, this paper utilizes two kinds of context information to address sparsity issue: the effect of hierarchical structure on users potential preferences and the dynamic effect of users short term ratings. An integrated model CICF is then proposed based on the two of the features mentioned above. Experimental results on Yahoo! Music ratings show that CICF could significantly improve the predication performance compared to baseline method. Furthermore, it is also demonstrated that CICF could effectively mitigate rating sparsity issue.
关键词
三协同过滤 /
上下文信息 /
隐参数模型
{{custom_keyword}} /
Key words
collaborative filtering /
context information /
latent factor model
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Badrul Sarwar, George Karypis, Joseph Konstan, et al. Item Based Collaborative Filtering Recommendation Algorithms[C]//Proceedings of WWW10, 2001: 285-295.
[2] Yahoo! Music dataset[DB/OL]. http://kddcup.yahoo.com/.
[3] FZhang, C H Y. A Collaborative Filtering Algorithm Embedded BP Network to Ameliorate Aparsity Issue[C]//Proceedings of International Conference on Machine Learning and Cybernetics. 2005.
[4] 赵琴琴,鲁凯,王斌.SPCF: 一种基于内存的传播式协同过滤推荐算法[C]//CCIR 2011.
[5] B M Sarwar, G Karypis, J A K, Riedl J. Application of Dimensionality Reduction in Recommender System Case Study[R]. 2000.
[6] Y Ding, X Li. Time weight collaborative filtering[C]//Proceedings of 14th ACM International Conference on Information and Knowledge Management (CIKM04), 2004: 485-492.
[7] Z Lu, D Agarwal, and S Dhillon. A Spatio-temporal Approach to Collaborative Filtering[C]//Proceedings of 3rd ACM Conference on Recommender Systems, RecSys09, NY, USA, 2009:13-20.
[8] Y Koren. Collaborative Filtering with Temporal Dynamics[C]//Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. KDD09, NY, 2009:89-97.
[9] M Montaner, B Lopez, Josep. A Taxonomy of Recommender Agents on the Internet[J]. Artif. Intell. Rev., 2003, 19:285-330.
[10] Lamere P. Social Tagging and Music Information Retrieval[J]. Journal of New Music Research, 2008,37(2): 101-114.
[11] Dror G, Koenigstein N, Koren Y. Yahoo! Music Recommendations: Modeling Music Ratings with Temporal Dynamics and Item Taxonomy[C]//Proceedings 5th ACM Conference on Recommender Systems.RecSys11, 2011.
[12] Toscher A, Jahrer M, Bell R. The Bigchaos Solution to the Netflix Grand Prize[R]. 2009.
[13] Koren Y. Factorization Meets the Neighborhood: A Multifaceted Collaborative Filtering Model[C]//Proceeding of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM 2008: 426-434.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61070111);中国科学院先导项目(XDA06030200)
{{custom_fund}}
PDF(2895 KB)
