多媒体信息网络相似度计算方法研究

南京大学学报(自然科学版) ›› 2015, Vol. 51 ›› Issue (2): 290–296.

多媒体信息网络相似度计算方法研究

施静静，张鹏，阮雅端，陈启美*

出版日期:2015-03-06 发布日期:2015-03-06
作者简介:(南京大学电子科学与工程学院,南京,210046)
基金资助:
国家科技重大专项( 2012ZX03005-004-003) , 国家自然科学基金(61105015)

Similarity measure research in multimediainformation networks

Shi Jingjing, Zhang Peng, RuanYaduan, Chen Qimei*

Online:2015-03-06 Published:2015-03-06
About author:(College of Electronic and Engineering, Nanjing University, Jiangsu, China210046)

摘要/Abstract

摘要： 在种类丰富、数据庞大的多媒体信息网络中，为兴趣相投的用户推荐相似视、音频文件已成为网络多媒体发展的亮点。为此，文中分析了结点相似度计算的成对随机游走传统模型，进而提出了一种相似度的新型计算方法ObjectSim。该算法结合结点属性信息和结点间基本关联信息，利用关联类型的加权原则，有效地提高了相似度计算的准确性。经规模数据集DBLP的实验表明，ObjectSim相似度计算准确性较传统图结构SimRank[1]算法有了显著提高。

Abstract: With the various and complicated multimedia information networks, recommending the similar video or audio for users who has similar interests, has become a highlight of the network multimedia development. So, this paper proposes a novel similarity measure ObjectSim based on pairwise random walks model. This algorithm combines the object link information with its partial attributes,and adds the link type weight set to improve measure accuracy efficiently. The experimental results on big data set DBLP verify the superiority compared with the SimRank, atraditional similarity measure based on graph

施静静，张鹏，阮雅端，陈启美*. 多媒体信息网络相似度计算方法研究[J]. 南京大学学报(自然科学版), 2015, 51(2): 290–296.

Shi Jingjing, Zhang Peng, RuanYaduan, Chen Qimei*. Similarity measure research in multimediainformation networks[J]. Journal of Nanjing University(Natural Sciences), 2015, 51(2): 290–296.

参考文献

[1]Jeh G, Widom J. SimRank: a measure of ¬¬str¬¬u¬ctural-context similarity [C] // Proceedings ofthe eighth ACM SIGKDD international conference on Knowledge discovery and data mining.ACM, 2002: 538-543.
[2]江敏,肖诗斌,王弘蔚,等.一种改进的基于《知网》的词语语义相似度计算[J].中文信息学报,2008,22(5):84-89.
[3]田久乐,赵蔚.基于同义词词林的词语相似度计算方法[J].吉林大学学报:信息科学版, 2010 (6): 602-608.
[4]石静,吴云芳,邱立坤,等.基于大规模语料库的汉语词义相似度计算方法[J].中文信息学报, 2013 (1): 1-6.
[5]尹坤,尹红风*,杨燕,等.基于 SimRank 的百度百科词条语义相似度计算[J].山东大学学报(工学版),2014,44(3):29-35.
[6]徐志明,李栋,刘挺,等.微博用户的相似性度量及其应用[J].计算机学报,2014,37(1): 207-218.
[7]马小军,赵伟.改进相似度的分布式个性化推荐[J].计算机工程与应用,2014(4):126-131.
[8]潘丽芳,杨炳儒.基于簇的K最近邻(KNN)分类算法研究[J].计算机工程与设计, 2009,(18).
[9]Titze I R. Nonlinear source–filter coupling in phonation: Theorya)[J]. The Journal of the Acoustical Society of America, 2008, 123.
[10] Couto T, Ziviani N, Calado P, et al. Classifying documents with link-based bibliometricmeasures[J]. Information Retrieval, 2010, 13:315-345.
[11] Jeh G, Widom J. Scaling personalized web se¬¬¬arch [C] //Proceedings of the 12th internatio¬¬¬¬¬¬¬¬¬¬¬¬nal conference on World Wide Web. ACM, 2003: 271-279.
[12]Balmin A, Hristidis V, Papakonstantinou Y. Objectrank: Authority-based keyword search in databases[C]//Proceedings of the Thirtieth international conference on Very large data bases-Volume 30. VLDB Endowment, 2004: 564-575.
[13] Nie Z, Zhang Y, Wen J R, et al. Object-level ranking: bringing order to web objects [C]//Proceedings of the 14th international conference on World Wide Web. ACM, 2005: 567-574.
[14] Xi W, Fox E A, Fan W, et al. Simfusion: measuring similarity using unified relationship matrix[C]//Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval. ACM, 2005: 130-137.
[15]Järvelin K, Kekäläinen J. Cumulated gain-based evaluation of IR techniques[J]. ACM Transactions on Information Systems (TOIS), 2002, 20(4): 422-446.
[16] Fogaras D, Rácz B. Scaling link-based simila¬r¬¬¬¬¬ity search[C]//Proceedings of the 14th international conference on World Wide Web. ACM, 2005: 641-650.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed