微博用户关注推荐及排名策略研究
|
摘要
随着互联网技术的迅速发展,微博作为一种新的社交网络服务,其应用越来越广,功能越来越丰富,影响力也越来越大。微博服务将用户与社交网络之间的距离进一步缩短,为用户提供便捷、快速的内容接收、发布方式。本文以当前国内较流行的新浪微博作为研究的数据来源,从微博与传统社交网络服务之间的差异入手,充分总结微博服务的新特性,并对两类用户指标进行分析。本文主要研究了微博服务中关注推荐和排名策略,如何为微博用户提供不同层次的关注对象以及对用户进行合理排名是本文研究重点。
首先,在微博服务中,用户可以自由关注其他用户,以获取这些用户的动态及微博信息,因此,为用户推荐高质量的关注对象是一种建立可靠用户关注关系,有效提高用户对微博服务依赖性的手段。本文以社会网络统计量为基础,提出一种适用于微博服务的新结构,并据此得到关注推荐模型。实验表明,该模型使用阈值对用户进行筛选后,两种子模型可以为用户提供不同亲密度的关注对象,有效增加了用户获取关注对象的途径。
另外,由于微博服务目标用户定位和发展模式的需要,目前普遍采用单一指标作为用户排名的依据。然而,单一指标排名方式难以真实地反映服务中的活跃用户。本文提出用户排名策略模型以计算用户的活跃指数。实验及分析表明,活跃指数在微博服务中能够较好地反映活跃用户的特征,适于度量用户活跃度排名情况。
With the rapid development of the Internet, the recent years have witnessed an explosion of microblog as a new hot killer SNS (Social network service) application. Actually, the microblog is more widely applied, more functional and more influence. It also narrows the gap between users and services and provides convenient information processing for microbloggers. This paper based on the data source from the Weibo.com, after analysing the distinction between microblog and traditional SNSs, summarizes some new features of the microblogging service. In addition, the analysis of the two types of users' indicators established the foundation for the subsequent research of this paper. The research works of this paper include following recommendation and ranking strategy.
Firstly, microblogging users can follow other users at will to receive their status and tweets. For establishing a reliable following relationship between the users and enhancing the users'reliance on the microblogging, it is an effective means to recommend high-quality following objects for user. This paper analyzes the differences between microblogging and other social network services and constructs a following recommendation model which fuses with the new configuration of social networks. Experiment result shows that the model uses thresholds to filter users firstly, and then two sub models provide following objects with different intimacy and raise the number of following effectively.
Additionally, because of its locating of targeted users and the need of the website's developing, monotonous index is widely used as the foundation of microbloggers'ranking. However, this type of ranking is hard to reflect the microbloggers'real activity in the service, since the differences between microblogging and other SNS (Social Networking Services). This paper focuses on two types of users'distributional feature and message propagation in Sina microblogging and Twitter. The model of how to compute the active index of microbloggers is also given. According to the experiments and analyzes, active index is capable of reflecting the characteristics of active microbloggers, which is suitable for the measurement of active microbloggers' ranking.
首先,在微博服务中,用户可以自由关注其他用户,以获取这些用户的动态及微博信息,因此,为用户推荐高质量的关注对象是一种建立可靠用户关注关系,有效提高用户对微博服务依赖性的手段。本文以社会网络统计量为基础,提出一种适用于微博服务的新结构,并据此得到关注推荐模型。实验表明,该模型使用阈值对用户进行筛选后,两种子模型可以为用户提供不同亲密度的关注对象,有效增加了用户获取关注对象的途径。
另外,由于微博服务目标用户定位和发展模式的需要,目前普遍采用单一指标作为用户排名的依据。然而,单一指标排名方式难以真实地反映服务中的活跃用户。本文提出用户排名策略模型以计算用户的活跃指数。实验及分析表明,活跃指数在微博服务中能够较好地反映活跃用户的特征,适于度量用户活跃度排名情况。
With the rapid development of the Internet, the recent years have witnessed an explosion of microblog as a new hot killer SNS (Social network service) application. Actually, the microblog is more widely applied, more functional and more influence. It also narrows the gap between users and services and provides convenient information processing for microbloggers. This paper based on the data source from the Weibo.com, after analysing the distinction between microblog and traditional SNSs, summarizes some new features of the microblogging service. In addition, the analysis of the two types of users' indicators established the foundation for the subsequent research of this paper. The research works of this paper include following recommendation and ranking strategy.
Firstly, microblogging users can follow other users at will to receive their status and tweets. For establishing a reliable following relationship between the users and enhancing the users'reliance on the microblogging, it is an effective means to recommend high-quality following objects for user. This paper analyzes the differences between microblogging and other social network services and constructs a following recommendation model which fuses with the new configuration of social networks. Experiment result shows that the model uses thresholds to filter users firstly, and then two sub models provide following objects with different intimacy and raise the number of following effectively.
Additionally, because of its locating of targeted users and the need of the website's developing, monotonous index is widely used as the foundation of microbloggers'ranking. However, this type of ranking is hard to reflect the microbloggers'real activity in the service, since the differences between microblogging and other SNS (Social Networking Services). This paper focuses on two types of users'distributional feature and message propagation in Sina microblogging and Twitter. The model of how to compute the active index of microbloggers is also given. According to the experiments and analyzes, active index is capable of reflecting the characteristics of active microbloggers, which is suitable for the measurement of active microbloggers' ranking.
引文
[1]中国互联网络信息中心.第29次中国互联网络发展状况统计报告[R].2012.http://www.cnnic.cn/research/bgxz/tjbg/201201/P020120118512855484817.pdf
[2]艾瑞网.Facebook值1000 亿美元吗? [EB/OL].2012.http://web2.iresearch.cn/59/20120220/163717.shtml
[3]Google. Google Trends[EB/OL].2012. http://www.google.cn/trends/
[4]Wikipedia.Microblogging[EB/OL].2012. http://en.wikipedia.org/wiki/Microblogging
[5]GigaTweet.Counting the number of tweets[EB/OL].2010. http://popacular.com/gigatweet
[6]新浪微博应用广场.应用首页[EB/OL].2012. http://weibo.com/app
[7]S. Milgram.The small world Problem[J]. Psychology Today,1967,2:60-67
[8]J. Guare. Six degrees of separation:A play[M]. Vintage Books, New York,1990.
[9]P. S. Dodds, R. Muhamad, and D. J. Watts. An experimental study of search in global social networks[J]. Science,2003,301(5634):827-829
[10]J. Leskovec and E. Horvitz. Planetary-scale views on a large instant-messaging network[C]. In:Proc. of the International Conference on World Wide Web,2008,915-924
[11]CNN Tech. On Facebook, it's now 4.74 degrees of separation[EB/OL].2011. http://articles.cnn.com/2011-11-22/tech/tech_social-media_facebook-six-degrees_1_facebo ok-age-facebook-users-social-network?_s=PM:TECH
[12]M. Granovetter. The strength of weak ties[J]. American Journal of Sociology,1973, 78(6):1360-1380
[13]M. Granovetter. The Strength of Weak Ties:A Network Theory Revisited[M]. Sociological Theory,1982,1:201-233
[14]R. Dunbar. Co-evolution of neocortical size, group size and language in humans[J]. Behavioral and Brain Sciences,1993,16(4):681-735
[15]M. Gladwell. The Tipping Point:How Little Things Can Make a Big Difference[M]. Little Brown & Company, New York,2000
[16]S.A. Golder, D. Wilkinson, and B.A. Huberma. Rhythms of social interaction:Messaging within a massive online network[C]. In:Proc. of the Third International Conference on Communities and Technologies,2007,41-66
[17]S.T. Tong, D.H. Brandon, L. Lwell, and J.B. Walther. Too much of a good thing? the relationship between number of friends and interpersonal impressions on facebook[J]. Journal of Computer-Mediated Communication,2008,13(3):531-549
[18]H. Kwak, C. Lee, H. Park and S. Moon. What is twitter, a social network or a news media? [C].In:Proc. of the International Conference on World Wide Web,2010,591-600
[19]M. Cha, H. Haddadi, F. Benevenuto and K. P. Gummadi. Measuring user influence in Twitter:The million follower fallacy[C]. In:Proc. of the Fourth International AAAI Conference on Weblogs and Social Media,2010,10-17
[20]T. Fujisaka, R. Lee, K. Sumiya. Detection of Unusually Crowded Places through Micro-Blogging Sites[C]. In the International Conference on Advanced Information Networking and Applications Workshops,2010,467-472
[21]Y. Shen, S. C. Li, X. X. Ye. Content Mining and Network Analysis of Microblog Spam [J]. Journal of Convergence Information Technology,2010,5(1):135-140
[22]L.J. Hong, Brian D. Davison. Empirical Study of Topic Modeling in Twitter[C]. In:Proc. of the SIGKDD Workshop on Social Media Analytics at KDD,2010
[23]C. Holotescu, G. Grosseck. Using microblogging in education Case Study:Cirip.ro[C]. In: Proc. of the Sixth International Conference on e-Learning,2010
[24]李华,赵文伟.微博客：图书馆的下一个网络新贵工具[J].图书与情报.2009,4：78-82
[25]B.W.Yang; W.C. Tsai; A.P. Chen; Ramandeep, S. Cloud Computing Architecture for Social Computing-A Comparison Study of Facebook and Google[C]. In the International Conference on Advances in Social Network Analysis and Mining,2011,741-745
[26]Merani, M. Luisa. How Helpful Can Social Network Friends Be in Peer-to-Peer Video Distribution?[C]. In the International Conference on Parallel and Distributed Systems, 2011,799-804
[27]和讯网.新新浪微博桌面加速升级 增强即时通讯能力[EB/OL].2011.http://tech.hexun.com/2011-11-14/135203847.html
[28]新浪微博.桌面客户端[EB/OL].2012. http://desktop.weibo.com/
[29]和讯网.YouTube与Google+加强整合提升视频共享社交[EB/OL].2012.http://tech.hexun.com/2012-02-19/138409526.html
[30]新浪科技Twitter新浪潮[EB/OL].2009. http://t.cn/h5Zhir
[31]Wikipedia.Sina Weibo[EB/OL].2012. http://en.wikipedia.org/wiki/Sina_Weibo
[32]Wikipedia.Renren[EB/OL].2012. http://en.wikipedia.org/wiki/Renren
[33]Wikipedia.Twitter[EB/OL].2012. http://en.wikipedia.org/wiki/Twitter
[34]新浪微博.姚晨的微博首页[EB/OL].2012. http://www.weibo.com/yaochen
[35]刘军.社会网络分析导论[M].北京：社会科学文献出版社,2004,45-56
[36]A. Barrat, M. Weigt. On the properties of small-world networks[J].European Physical Journal B,2000,13(3):547-560
[37]H.J. Kim, J.M. Kim. Cyclic topology in complex network[J]. physics/0503168,2005
[38]D.J. Watts and S.H. Strogatz. Collective dynamics of'small-world'networks [J].Nature, 1998,393(6684):440-442
[39]T. Zhou, G. Yan and B. H. Wang. Maximal planar networks with large clustering coefficient and power-law degree distribution[J]. Phys. Rev. E,2005,71:046141.1-046141.12
[40]A.L. Barabasi, R. Albert. Emergence of scaling in random networks[J]. Science,1999,286: 509-512
[41]赵丽,袁睿翕,管晓宏等.博客网络中具有突发性的话题传播模型[J].软件学报,2009,20(5)：1384-1392
[42]J. Vitak, N.B. Ellison and C. Steinfield. The Ties That Bond:Re-Examining the Relationship between Facebook Use and Bonding Social Capital[C]. In the Hawaii International Conference on System Sciences,2011,1-10
[43]T. Snijders, A.B. Tom, E. P. Philippa and L. R. Garry. New Specifications for Exponential Random Graph Models[J]. Sociological Methodology,2006,36(1):99-153
[44]G.L. Robins, P.E. Pattison, Y. Kalish and D. Lusher. An introduction to exponential random graph (p*) models for social networks[J]. Social Networks,2007,29:173-191
[45]G.L. Robins, T. Snijders, P. Wang, M. Handcock and P.E. Pattison. Closure, Connectivity and Degree Distributions:Exponential Random Graph (p*) Models for Directed Social Networks[J]. Social Networks,2009,31:105-117
[46]X. Xie. Potential Friend Recommendation in Online Social Network[C]. In:Green Computing and Communications,2010 IEEE/ACM Int'l Conference on & International Conference on Cyber, Physical and Social Computing,2010,831-835
[47]S.L. Kang, C. Zhang, Z.Q. Lin, X.M. Shi, H. Ma. Complexity Research of Massively Microblogging Based on Human Behaviors[C]. In:Proc. of Database Trends and Applications,2010,1-4
[48]X. Yu, A. Pan, L.A. Tang, Z.H. Li, J.W. Han. Geo-Friends Recommendation in GPS-based Cyber-physical Social Network[C]. In:Proc. of the International Conference on Advances in Social Network Analysis and Mining,2011,361-368
[49]D. Lee, S.E. Park, M. Kahng, S. Lee and S. G. Lee. Exploiting contextual information from event logs for personalized recommendation[C]. In:Proc. of the Computer and Information Science 2010:Studies in Computational Intelligence, Springer Berlin/Heidelberg,2010,317: 121-139
[50]T.L. Saaty. Theory and Applications of the Analytic Network Process[M]. RWS Publications,2005
[51]T.L. Saaty. Decision making with the analytic hierarchy process[J]. International Journal of Services Sciences,2008,1(1):83-98
[52]T.L. Saaty. The Analytic Hierarchy Process[M].New York:McGraw-Hill 1980
[53]姜启源.数学模型(第二版)[M].北京,高等教育出版社,1993,305-336
[54]A. Ritter, C. Cherry and B. Dolan. Unsupervised modeling of Twitter conversations[C]. In: Proc. of the North American Chapter of the Association for Computational Linguistics,2010,172-180
[55]J. Tang, J. Sun, C. Wang and Z. Yang. Social influence analysis in large-scale networks[C]. In:Proc. of the Knowledge Discovery and Data Mining,2009,807-816
[56]J.S. Weng, E.P. Lim, J. Jiang, et al. TwitterRank:finding topic-sensitive influential twitterers[C]. In:Proc. of the Third ACM International Conference on Web Search and Data Mining,2010,261-270
[2]艾瑞网.Facebook值1000 亿美元吗? [EB/OL].2012.http://web2.iresearch.cn/59/20120220/163717.shtml
[3]Google. Google Trends[EB/OL].2012. http://www.google.cn/trends/
[4]Wikipedia.Microblogging[EB/OL].2012. http://en.wikipedia.org/wiki/Microblogging
[5]GigaTweet.Counting the number of tweets[EB/OL].2010. http://popacular.com/gigatweet
[6]新浪微博应用广场.应用首页[EB/OL].2012. http://weibo.com/app
[7]S. Milgram.The small world Problem[J]. Psychology Today,1967,2:60-67
[8]J. Guare. Six degrees of separation:A play[M]. Vintage Books, New York,1990.
[9]P. S. Dodds, R. Muhamad, and D. J. Watts. An experimental study of search in global social networks[J]. Science,2003,301(5634):827-829
[10]J. Leskovec and E. Horvitz. Planetary-scale views on a large instant-messaging network[C]. In:Proc. of the International Conference on World Wide Web,2008,915-924
[11]CNN Tech. On Facebook, it's now 4.74 degrees of separation[EB/OL].2011. http://articles.cnn.com/2011-11-22/tech/tech_social-media_facebook-six-degrees_1_facebo ok-age-facebook-users-social-network?_s=PM:TECH
[12]M. Granovetter. The strength of weak ties[J]. American Journal of Sociology,1973, 78(6):1360-1380
[13]M. Granovetter. The Strength of Weak Ties:A Network Theory Revisited[M]. Sociological Theory,1982,1:201-233
[14]R. Dunbar. Co-evolution of neocortical size, group size and language in humans[J]. Behavioral and Brain Sciences,1993,16(4):681-735
[15]M. Gladwell. The Tipping Point:How Little Things Can Make a Big Difference[M]. Little Brown & Company, New York,2000
[16]S.A. Golder, D. Wilkinson, and B.A. Huberma. Rhythms of social interaction:Messaging within a massive online network[C]. In:Proc. of the Third International Conference on Communities and Technologies,2007,41-66
[17]S.T. Tong, D.H. Brandon, L. Lwell, and J.B. Walther. Too much of a good thing? the relationship between number of friends and interpersonal impressions on facebook[J]. Journal of Computer-Mediated Communication,2008,13(3):531-549
[18]H. Kwak, C. Lee, H. Park and S. Moon. What is twitter, a social network or a news media? [C].In:Proc. of the International Conference on World Wide Web,2010,591-600
[19]M. Cha, H. Haddadi, F. Benevenuto and K. P. Gummadi. Measuring user influence in Twitter:The million follower fallacy[C]. In:Proc. of the Fourth International AAAI Conference on Weblogs and Social Media,2010,10-17
[20]T. Fujisaka, R. Lee, K. Sumiya. Detection of Unusually Crowded Places through Micro-Blogging Sites[C]. In the International Conference on Advanced Information Networking and Applications Workshops,2010,467-472
[21]Y. Shen, S. C. Li, X. X. Ye. Content Mining and Network Analysis of Microblog Spam [J]. Journal of Convergence Information Technology,2010,5(1):135-140
[22]L.J. Hong, Brian D. Davison. Empirical Study of Topic Modeling in Twitter[C]. In:Proc. of the SIGKDD Workshop on Social Media Analytics at KDD,2010
[23]C. Holotescu, G. Grosseck. Using microblogging in education Case Study:Cirip.ro[C]. In: Proc. of the Sixth International Conference on e-Learning,2010
[24]李华,赵文伟.微博客：图书馆的下一个网络新贵工具[J].图书与情报.2009,4：78-82
[25]B.W.Yang; W.C. Tsai; A.P. Chen; Ramandeep, S. Cloud Computing Architecture for Social Computing-A Comparison Study of Facebook and Google[C]. In the International Conference on Advances in Social Network Analysis and Mining,2011,741-745
[26]Merani, M. Luisa. How Helpful Can Social Network Friends Be in Peer-to-Peer Video Distribution?[C]. In the International Conference on Parallel and Distributed Systems, 2011,799-804
[27]和讯网.新新浪微博桌面加速升级 增强即时通讯能力[EB/OL].2011.http://tech.hexun.com/2011-11-14/135203847.html
[28]新浪微博.桌面客户端[EB/OL].2012. http://desktop.weibo.com/
[29]和讯网.YouTube与Google+加强整合提升视频共享社交[EB/OL].2012.http://tech.hexun.com/2012-02-19/138409526.html
[30]新浪科技Twitter新浪潮[EB/OL].2009. http://t.cn/h5Zhir
[31]Wikipedia.Sina Weibo[EB/OL].2012. http://en.wikipedia.org/wiki/Sina_Weibo
[32]Wikipedia.Renren[EB/OL].2012. http://en.wikipedia.org/wiki/Renren
[33]Wikipedia.Twitter[EB/OL].2012. http://en.wikipedia.org/wiki/Twitter
[34]新浪微博.姚晨的微博首页[EB/OL].2012. http://www.weibo.com/yaochen
[35]刘军.社会网络分析导论[M].北京：社会科学文献出版社,2004,45-56
[36]A. Barrat, M. Weigt. On the properties of small-world networks[J].European Physical Journal B,2000,13(3):547-560
[37]H.J. Kim, J.M. Kim. Cyclic topology in complex network[J]. physics/0503168,2005
[38]D.J. Watts and S.H. Strogatz. Collective dynamics of'small-world'networks [J].Nature, 1998,393(6684):440-442
[39]T. Zhou, G. Yan and B. H. Wang. Maximal planar networks with large clustering coefficient and power-law degree distribution[J]. Phys. Rev. E,2005,71:046141.1-046141.12
[40]A.L. Barabasi, R. Albert. Emergence of scaling in random networks[J]. Science,1999,286: 509-512
[41]赵丽,袁睿翕,管晓宏等.博客网络中具有突发性的话题传播模型[J].软件学报,2009,20(5)：1384-1392
[42]J. Vitak, N.B. Ellison and C. Steinfield. The Ties That Bond:Re-Examining the Relationship between Facebook Use and Bonding Social Capital[C]. In the Hawaii International Conference on System Sciences,2011,1-10
[43]T. Snijders, A.B. Tom, E. P. Philippa and L. R. Garry. New Specifications for Exponential Random Graph Models[J]. Sociological Methodology,2006,36(1):99-153
[44]G.L. Robins, P.E. Pattison, Y. Kalish and D. Lusher. An introduction to exponential random graph (p*) models for social networks[J]. Social Networks,2007,29:173-191
[45]G.L. Robins, T. Snijders, P. Wang, M. Handcock and P.E. Pattison. Closure, Connectivity and Degree Distributions:Exponential Random Graph (p*) Models for Directed Social Networks[J]. Social Networks,2009,31:105-117
[46]X. Xie. Potential Friend Recommendation in Online Social Network[C]. In:Green Computing and Communications,2010 IEEE/ACM Int'l Conference on & International Conference on Cyber, Physical and Social Computing,2010,831-835
[47]S.L. Kang, C. Zhang, Z.Q. Lin, X.M. Shi, H. Ma. Complexity Research of Massively Microblogging Based on Human Behaviors[C]. In:Proc. of Database Trends and Applications,2010,1-4
[48]X. Yu, A. Pan, L.A. Tang, Z.H. Li, J.W. Han. Geo-Friends Recommendation in GPS-based Cyber-physical Social Network[C]. In:Proc. of the International Conference on Advances in Social Network Analysis and Mining,2011,361-368
[49]D. Lee, S.E. Park, M. Kahng, S. Lee and S. G. Lee. Exploiting contextual information from event logs for personalized recommendation[C]. In:Proc. of the Computer and Information Science 2010:Studies in Computational Intelligence, Springer Berlin/Heidelberg,2010,317: 121-139
[50]T.L. Saaty. Theory and Applications of the Analytic Network Process[M]. RWS Publications,2005
[51]T.L. Saaty. Decision making with the analytic hierarchy process[J]. International Journal of Services Sciences,2008,1(1):83-98
[52]T.L. Saaty. The Analytic Hierarchy Process[M].New York:McGraw-Hill 1980
[53]姜启源.数学模型(第二版)[M].北京,高等教育出版社,1993,305-336
[54]A. Ritter, C. Cherry and B. Dolan. Unsupervised modeling of Twitter conversations[C]. In: Proc. of the North American Chapter of the Association for Computational Linguistics,2010,172-180
[55]J. Tang, J. Sun, C. Wang and Z. Yang. Social influence analysis in large-scale networks[C]. In:Proc. of the Knowledge Discovery and Data Mining,2009,807-816
[56]J.S. Weng, E.P. Lim, J. Jiang, et al. TwitterRank:finding topic-sensitive influential twitterers[C]. In:Proc. of the Third ACM International Conference on Web Search and Data Mining,2010,261-270