›› 2012,Vol. 27 ›› Issue (6): 1252-1260.doi: 10.1007/s11390-012-1301-5

• Special Section on Selected Paper from NPC 2011 • 上一篇    下一篇

JacUOD:协同过滤中一种新的相似性度量方法

Hui-Feng Sun1 (孙慧峰), Student Member, CCF, ACM, Jun-Liang Chen1 (陈俊亮), Gang Yu1 (俞钢), Member, IEEE, Chuan-Chang Liu1 (刘传昌), Member, IEEE, Yong Peng1 (彭泳), Guang Chen2 (陈光), and Bo Cheng1 (程渤), Senior Member, CCF, Member, ACM   

  • 收稿日期:2011-08-23 修回日期:2012-01-19 出版日期:2012-11-05 发布日期:2012-11-05

JacUOD: A New Similarity Measurement for Collaborative Filtering

Hui-Feng Sun1 (孙慧峰), Student Member, CCF, ACM, Jun-Liang Chen1 (陈俊亮), Gang Yu1 (俞钢), Member, IEEE, Chuan-Chang Liu1 (刘传昌), Member, IEEE, Yong Peng1 (彭泳), Guang Chen2 (陈光), and Bo Cheng1 (程渤), Senior Member, CCF, Member, ACM   

  1. 1. State Key Lab of Network and Switching Technology, Beijing University of Posts and Telecommunications Beijing 100876, China;
    2. School of Information and Communication Engineering, Beijing University of Posts and Telecommunications Beijing 100876, China
  • Received:2011-08-23 Revised:2012-01-19 Online:2012-11-05 Published:2012-11-05
  • Supported by:

    This work was supported by the National Basic Research 973 Program of China under Grant No. 2011CB302506, the National Natural Science Foundation of China under Grant Nos. 61001118, 61132001, 61003067, the National Major Science and Technology Project of New Generation Broadband Wireless Network of China under Grant No. 2010ZX03004-001, and the Fundamental Research Funds for the Central Universities of Beijing University of Posts and Telecommunications of China under Grant No. 2011RC0502.

协同过滤已被广泛应用于推荐系统中,它能帮助用户发现自己喜欢的物品.对于协同过滤来说,相似性度量至关重要,它用来度量用户之间或物品之间的相似程度.然而,协同过滤的相似性度量方法仍有待提升.在这篇论文里,我们提出一种新的相似性度量方法JacUOD,用以有效地衡量相似度.JacUOD为不同维数向量空间里的向量之间进行相似度比较提供了统一的方法.与传统的相似性度量方法相比,JacUOD很好地处理了不同向量空间的维数差异问题.我们以基于用户的协同过滤为例,在电影评分数据集MovieLens上做了实验,以验证JacUOD的有效性.实验结果表明,我们的方法JacUOD比传统方法获得了更好的预测精度.

Abstract: Collaborative filtering (CF) has been widely applied to recommender systems, since it can assist users to discover their favorite items. Similarity measurement that measures the similarity between two users or items is critical to CF. However, traditional similarity measurement approaches for memory-based CF can be strongly improved. In this paper, we propose a novel similarity measurement, named Jaccard Uniform Operator Distance (JacUOD), to effectively measure the similarity. Our JacUOD approach aims at unifying similarity comparison for vectors in different multidimensional vector spaces. Compared with traditional similarity measurement approaches, JacUOD properly handles dimension-number difference for different vector spaces. We conduct experiments based on the well-known MovieLens datasets, and take user-based CF as an example to show the effectiveness of our approach. The experimental results show that our JacUOD approach achieves better prediction accuracy than traditional similarity measurement approaches.

中图分类号: 

  • null
[1] Balabanovic M, Shoham Y. Fab: Content-based collaborativerecommendation. Comm. ACM, 1997, 40(3): 66-72.

[2] Lang K. NewsWeeder: Learning to filter netnews. In Proc.the 12th Int. Conf. Machine Learning, Jul. 1995, pp.331-339.

[3] Mooney R J, Roy L. Content-based book recommending usinglearning for text categorization. In Proc. ACM SIGIR 1999Workshop Recommender Systems: Algorithms and Evalua-tion, Aug. 1999, pp.195-204.

[4] Pazzani M, Billsus D. Learning and revising user profiles:The identification of interesting web sites. Machine Learn-ing, 1997, 27(3): 313-331.

[5] Xue G R, Dai W Y, Yang Q, Yu Y. Topic-bridged PLSA forcross-domain text classification. In Proc. the 31st Conf. Re-search and Development in Information Retrieval, Jul. 2008,pp.627-634.

[6] Liu N N, Yang Q. EigenRank: A ranking-oriented approach tocollaborative filtering. In Proc. the 31st Conf. Research andDevelopment in Information Retrieval, Jul. 2008, pp.83-90.

[7] Breese J, Heckerman D, Kadie C. Empirical analysis of predic-tive algorithms for collaborative filtering. In Proc. the 14thInt. Conf. Uncertainty in Artificial Intelligence (UAI 1998),May 1998, pp.43-52.

[8] Liu N N, Zhao M, Yang Q. Probabilistic latent prefer-ence analysis for collaborative filtering. In Proc. the18th Int. Conf. Information and Knowledge Management(CIKM2009), Nov. 2009, pp.759-766.

[9] Xin X, King L, Deng H, Lyu M R. A social recommendationframework based on multi-scale continuous conditional ran-dom fields. In Proc. the 18th ACM Conf. Information andKnowledge Management, Nov. 2009, pp.1247-1256.

[10] Adomavicius G, Tuzhilin A. Toward the next generation ofrecommender systems: A survey of the state-of-the-art andpossible extensions. IEEE Trans. Knowledge and Data En-gineering, 2005, 17(6): 734-749.

[11] Herlocker J L, Konstan J A, Borchers A, Riedl J. An algo-rithmic framework for performing collaborative filtering. InProc. the 22nd Conf. Research and Development in Infor-mation Retrieval, Aug. 1999, pp.230-237.

[12] Jin R, Chai J Y, Si L. An automatic weighting scheme forcollaborative filtering. In Proc. the 27th Conf. Research andDevelopment in Information Retrieval, Jul. 2004, pp.337-344.

[13] Zhao Z, Shang M. User-based collaborative-filtering recom-mendation algorithms on Hadoop. In Proc. the 3rd Int.Conf. Knowledge Discovery and Data Mining, Jan. 2010,pp478-481.

[14] Deshpande M, Karypis G. Item-based top-N recommenda-tion algorithms. ACM Transactions on Information Systems,2004, 22(1): 143-177.

[15] Linden G, Smith B, York J. Amazon.com recommendations:Item-to-item collaborative filtering. IEEE Internet comput-ing, 2003, 7(1): 76-80.

[16] Sarwar B, Karypis G, Konstan J, Reidl J. Item-based collabo-rative filtering recommendation algorithms. In Proc. the 10thInt. World Wide Web Conference, May 2001, pp.285-295.

[17] Ma H, King I, Lyu M. Effective missing data prediction forcollaborative filtering. In Proc. the 30th Conf. Research andDevelopment in Information Retrieval, Jul. 2007, pp.39-46.

[18] Wang J, Vries De A, Reinders M. Unifying user-based anditem-based collaborative filtering approaches by similarity fu-sion. In Proc. the 29th Conf. Research and Development inInformation Retrieval, Aug. 2006, pp.501-508.

[19] Zheng Z, Ma H, Lyu M R, King I. WSRec: A collaborativefiltering based Web service recommender system. In Proc.the 7th Int. Conf. Web Services, Jul. 2009, pp.437-444.

[20] Zheng Z, Ma H, Lyu M R, King I. QoS-awareWeb service rec-ommendation by collaborative filtering. IEEE Transactionson Service Computing, 2011, 4(2): 140-152.

[21] Resnick P, Iacovou N, Sushak M, Bergstrom P, Riedl J. Grou-pLens: An open architecture for collaborative filtering of net-news. In Proc. the 1994 ACM Conf. Computer SupportedCooperative Work, Oct. 1994, pp.175-186.

[22] Shardanand U, Maes P. Social information filtering: Algo-rithms for automating "Word of Mouth". In Proc. HumanFactors in Computing Systems, May 1995, pp.210-217.

[23] Candillier L, Meyer F, Fessant F. Designing specific weightedsimilarity measures to improve collaborative filtering systems.In Proc. the 12th Industrial Conf. Data Mining, Jul. 2008,pp.242-255.

[24] Ahn J H. A new similarity measure for collaborative filteringto alleviate the new user cold-starting problem. InformationSciences, 2008, 178(1): 37-51.

[25] Zeng C, Xing C X, Zhou L Z, Zheng X H. Similarity measureand instance selection for collaborative filtering. Int. Journalof Electronic Commerce, 2004, 8(4): 115-129.

[26] Fouss F, Pirotte A, Renders J M, Saerens M. Random-walkcomputation of similarities between nodes of a graph withapplication to collaborative recommendation. IEEE Trans.Knowledge and Data Engineering, 2007, 19(3): 355-369.

[27] Symeonidis P, Nanopoulos A, Papadopoulos A N, Manolopou-los Y. Collaborative filtering: Fallacies and insights in mea-suring similarity. In Proc. the 10th PKDD Workshop on WebMining, Sept. 2006, pp.56-67.

[28] Shi Y, Larson M, Hanjalic A. Exploiting user similarity basedon rated-item pools for improved user-based collaborative fil-tering. In Proc. the 3rd ACM Conf. Recommender Systems,Oct. 2009, pp.125-132.

[29] Miller B, Albert I, Lam S, Konstan J, Riedl J. MovieLensunplugged: Experiences with an occasionally connected rec-ommender system. In Proc. the 8th International Conferenceon Intelligent User Interfaces, Jan. 2003, pp.263-266
[1] Ying-Jun Wu1 (吴英骏), Han Huang1,2,* (黄翰), Member, CCF, ACM, IEEE, Zhi-Feng Hao3 (郝志峰), and Feng Chen4 (陈丰). 基于链接相似度的局部社团发现[J]. , 2012, 27(6): 1261-1268.
[2] Punam Bedi1, Member, ACM, Senior Member, IEEE, Anjali Thukral2,*, Hema Banati3, Abhishek Behl1,**, and Varun Mendiratta1,**. 一种多线程语义聚焦的爬虫[J]. , 2012, 27(6): 1233-1242.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 高庆狮; 张祥; 杨树范; 陈树清;. Vector Computer 757[J]. , 1986, 1(3): 1 -14 .
[2] 韩建超; 史忠植;. Formalizing Default Reasoning[J]. , 1990, 5(4): 374 -378 .
[3] 张钹; 张铃;. On Memory Capacity of the Probabilistic Logic Neuron Network[J]. , 1993, 8(3): 62 -66 .
[4] 罗军舟; 顾冠群;. CIMS Network Protocol and Its Net Models[J]. , 1997, 12(5): 476 -481 .
[5] 万华根; 金小刚; 鲍虎军;. Direct 3D Painting with a Metaball-Based Paint brush[J]. , 2000, 15(1): 100 -104 .
[6] 章文嵩; 金士尧; 吴泉源;. LinuxDirector: A Connection Director for Scalable Internet Services[J]. , 2000, 15(6): 560 -571 .
[7] . 颜色图像滤波的四元数扩散方法[J]. , 2006, 21(1): 126 -136 .
[8] . 从监督视角进行多示例学习[J]. , 2006, 21(5): 800 -809 .
[9] . 使用应变场插值的三维图形渐变[J]. , 2007, 22(1): 147 -155 .
[10] . 暂缺[J]. , 2007, 22(3): 371 -378 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn
总访问量: