影响力扩散概率模型及其用于意见领袖发现研究
|
摘要
作为意见领袖识别基础的影响力扩散模型IDM存在两个缺陷:(1)由回复链结构断层或者帖子内容间接传播引起的影响力传递中断;(2)由灌水所导致的虚假影响力传递.为解决上述问题,文中提出了一种新的影响力扩散概率模型IDPM,进而建立了网络意见领袖筛选模型.该模型在相同兴趣空间上定义单个关键词语传播概率影响力,在帖子影响力定义中引入了有效关键词语概念,避免了上述缺陷;同时,在用户影响力计算时给每个帖子一个影响因子,用以整合其它有用信息,使模型具有开放性和包容性特点.在2010年12月到2011年5月网易社会新闻版块评论数据上的实验表明,文中方法是有效的,其平均精确率相对IDM模型提高了59.8%.
There exist two defects in the influence diffusion model,which is a base for opinion leader identification.One is the influence diffusion break caused by the broken reply chain or indirect content diffusion,and the other is illusive influence diffusion caused by flooding posts.To solve the above problems,this paper presents a new Influence Diffusion Probability Model(IDPM),and then builts a network opinion leader identification model.In which,the diffusion probability influence of the single term is defined in the same interesting space,and the concept of valid term in post influence evaluation is introduced.An impact fact to be added to influence calculation,which integrates other useful information,and this leads the model is open and inclusive.The experiments in the data collected from the NetEase social news section from December 2010 to May 2011,show that the method proposed in this paper is valid,and the average precision is more 59.8% than that of IDM.
There exist two defects in the influence diffusion model,which is a base for opinion leader identification.One is the influence diffusion break caused by the broken reply chain or indirect content diffusion,and the other is illusive influence diffusion caused by flooding posts.To solve the above problems,this paper presents a new Influence Diffusion Probability Model(IDPM),and then builts a network opinion leader identification model.In which,the diffusion probability influence of the single term is defined in the same interesting space,and the concept of valid term in post influence evaluation is introduced.An impact fact to be added to influence calculation,which integrates other useful information,and this leads the model is open and inclusive.The experiments in the data collected from the NetEase social news section from December 2010 to May 2011,show that the method proposed in this paper is valid,and the average precision is more 59.8% than that of IDM.
引文
[1]Liu Jian-Ming.Public Opinion Propagation.Beijing:TsinghuaUniversity Press,2001(in Chinese)(刘建明.舆论传播.北京:清华大学出版社,2001)
[2]Lazarsfield P et al.The People’s Choice.New York:Columbia University Press,1948
[3]Weimann Gabriel,Tustin Deon Harold,van Vuuren Daan,Joubert J P R.Looking for opinion leaders:Traditional vs.modern measures in traditional societies.InternationalJournal of Public Opinion Research,2007,19(2):173-190
[4]Matsumura Naohiro,Ohsawa Yukio,Ishizuka Mitsuru.Influence diffusion model in text-based communication.Transactions of the Japanese Society for Artificial Intelli-gence,2002,17(3):259-267
[5]Yu Hong.Research on the opinion leaders of political BBS:An case study on Sino-Japan BBS of strong nation forum[Ph.D.dissertation].Huazhong University of Science andTechnology,Wuhan,2007:32-36(in Chinese)(余红.网络时政论坛舆论领袖模型初探[博士学位论文].华中科技大学,武汉,2007:32-36)
[6]Shi Mao,Fang Yong,Zeng Xiang-Ping,Wang Chang-Hui.Analysis of IDM model and its reformative arithmetic.Jour-nal of Chengdu University of Information Technology,2008,23(1):69-72(in Chinese)(石矛,方勇,曾祥平,王长辉.IDM模型分析及其影响力改进算法.成都信息工程学院学报,2008,23(1):69-72)
[7]Zhou Hengmin,Zeng Daniel,Zhang Changli.Findingleaders from opinion networks//Proceedings of the 2009IEEE International Conference on Intelligence and SecurityInformatics(ISI’09).Dallas TX,USA,2009:266-268
[8]Gao Jun-Bo,Yang Jing.Analysis of opinion leader in on-linecommunities.Journal of University of Electronic Science andTechnology of China,2007,36(6):1249-1252(in Chinese)(高俊波,杨静.在线论坛的意见领袖分析.电子科技大学学报,2007,36(6):1249-1252)
[9]Zhang Jun,Ackerman Mark S,Adamic Lada.Expertise net-works in online communities:Structure and algorithms//Proceedings of the 16th International World Wide Web Con-ference Committee(IW3C2).Banff,Canada,2007:221-230
[10]Song Xiaodan,Chi Yun,Hino Koji,Belle Tseng.Identifyingopinion leaders in the blogosphere//Proceedings of the 16thACM Conference on Information and Knowledge Manage-ment(CIKM’07).New York,USA,2007:971-974
[11]Zhai Zhongwu,Xu Hua,Jia Peifa.Identifying opinion lead-ers in BBS//Proceedings of the IEEE/WIC/ACM Interna-tional Conference on Web Intelligence and Intelligent AgentTechnology(WI-IAT’08).Sydney,NSW,Australia,2008:398-401
[12]Wang Jue,Zeng Jian-Ping,Zhou Bao-Hua,Wu Cheng-Rong.Online forum opinion leaders discovering methodbased on clustering analysis.Computer Engineering,2011,37(5):44-46(in Chinese)(王珏,曾剑平,周葆华,吴承荣.基于聚类分析的网络论坛意见领袖发现方法.计算机工程,2011,37(5):44-46)
[13]Fan Xing-Hua,Sun Mao-Song.A high performance two-class Chinese text categorization method.Chinese Journal ofComputers,2006,29(1):124-131(in Chinese)(樊兴华,孙茂松.一种高性能的两类中文文本分类方法.计算机学报,2006,29(1):124-131)
[14]Fan Xing-Hua,Nie Jian-Yun.Link distribution dependencymodel for document retrieval.Journal of Information&Computational Science,2009,6(3):1553-1564
[2]Lazarsfield P et al.The People’s Choice.New York:Columbia University Press,1948
[3]Weimann Gabriel,Tustin Deon Harold,van Vuuren Daan,Joubert J P R.Looking for opinion leaders:Traditional vs.modern measures in traditional societies.InternationalJournal of Public Opinion Research,2007,19(2):173-190
[4]Matsumura Naohiro,Ohsawa Yukio,Ishizuka Mitsuru.Influence diffusion model in text-based communication.Transactions of the Japanese Society for Artificial Intelli-gence,2002,17(3):259-267
[5]Yu Hong.Research on the opinion leaders of political BBS:An case study on Sino-Japan BBS of strong nation forum[Ph.D.dissertation].Huazhong University of Science andTechnology,Wuhan,2007:32-36(in Chinese)(余红.网络时政论坛舆论领袖模型初探[博士学位论文].华中科技大学,武汉,2007:32-36)
[6]Shi Mao,Fang Yong,Zeng Xiang-Ping,Wang Chang-Hui.Analysis of IDM model and its reformative arithmetic.Jour-nal of Chengdu University of Information Technology,2008,23(1):69-72(in Chinese)(石矛,方勇,曾祥平,王长辉.IDM模型分析及其影响力改进算法.成都信息工程学院学报,2008,23(1):69-72)
[7]Zhou Hengmin,Zeng Daniel,Zhang Changli.Findingleaders from opinion networks//Proceedings of the 2009IEEE International Conference on Intelligence and SecurityInformatics(ISI’09).Dallas TX,USA,2009:266-268
[8]Gao Jun-Bo,Yang Jing.Analysis of opinion leader in on-linecommunities.Journal of University of Electronic Science andTechnology of China,2007,36(6):1249-1252(in Chinese)(高俊波,杨静.在线论坛的意见领袖分析.电子科技大学学报,2007,36(6):1249-1252)
[9]Zhang Jun,Ackerman Mark S,Adamic Lada.Expertise net-works in online communities:Structure and algorithms//Proceedings of the 16th International World Wide Web Con-ference Committee(IW3C2).Banff,Canada,2007:221-230
[10]Song Xiaodan,Chi Yun,Hino Koji,Belle Tseng.Identifyingopinion leaders in the blogosphere//Proceedings of the 16thACM Conference on Information and Knowledge Manage-ment(CIKM’07).New York,USA,2007:971-974
[11]Zhai Zhongwu,Xu Hua,Jia Peifa.Identifying opinion lead-ers in BBS//Proceedings of the IEEE/WIC/ACM Interna-tional Conference on Web Intelligence and Intelligent AgentTechnology(WI-IAT’08).Sydney,NSW,Australia,2008:398-401
[12]Wang Jue,Zeng Jian-Ping,Zhou Bao-Hua,Wu Cheng-Rong.Online forum opinion leaders discovering methodbased on clustering analysis.Computer Engineering,2011,37(5):44-46(in Chinese)(王珏,曾剑平,周葆华,吴承荣.基于聚类分析的网络论坛意见领袖发现方法.计算机工程,2011,37(5):44-46)
[13]Fan Xing-Hua,Sun Mao-Song.A high performance two-class Chinese text categorization method.Chinese Journal ofComputers,2006,29(1):124-131(in Chinese)(樊兴华,孙茂松.一种高性能的两类中文文本分类方法.计算机学报,2006,29(1):124-131)
[14]Fan Xing-Hua,Nie Jian-Yun.Link distribution dependencymodel for document retrieval.Journal of Information&Computational Science,2009,6(3):1553-1564