一种基于最大熵原理的社交网络用户关系分析模型

肖云鹏; 杨光; 刘宴兵; 吴斌

doi:10.11999/JEIT160605

一种基于最大熵原理的社交网络用户关系分析模型

doi: 10.11999/JEIT160605

1.
(重庆邮电大学网络与信息安全技术重庆市工程实验室重庆 400065) ②(北京邮电大学北京市智能通信软件与多媒体重点实验室北京 100876)

基金项目:

国家973计划项目(2013CB329606), 国家自然科学基金(61272400), 重庆市青年人才项目(cstc2013kjrc-qnrc 40004)，教育部-中国移动研究基金(MCM20130351)，重庆市研究生研究与创新项目(CYS14146)，重庆市教委科学计划项目(KJ1500425)，重庆邮电大学文峰基金(WF201403)

计量
- 文章访问数: 1477
- HTML全文浏览量: 153
- PDF下载量: 555
- 被引次数: 0
出版历程
- 收稿日期: 2016-06-07
- 修回日期: 2016-11-30
- 刊出日期: 2017-04-19

Social Relationship Analysis Model Based onthe Principle of Maximum Entropy

1.
(Chongqing Engineering Laboratory of Internet and Information Security, Chongqing University of Posts and Telecommunications, Chongqing 400065, China)
2.
(The Intelligent Communication Software and Multimedia Key Laboratory of Beijing, Beijing University of Posts and Telecommunications, Beijing 100876, China)

Funds:

The National 973 Program of China (2013CB 329606), The National Natural Science Foundation of China (61272400), Chongqing Youth Innovative Talent Project (cstc2013 kjrc-qnrc40004), Ministry of Education of China and China Mobile Research Fund (MCM20130351), Chongqing Graduate Research and Innovation Project (CYS14146), Science and Technology Research Program of the Chongqing Municipal Education Committee (KJ1500425), WenFeng Foundation of CQUPT (WF201403)

摘要

摘要: 在社交网络的演化和发展过程中，用户之间关系的建立受到多种因素的共同作用。该文通过对社交网络中用户属性以及用户关系数据进行分析，旨在发现影响用户关系建立的关键因素。首先，针对用户关系建立的复杂驱动因素，分别从个人兴趣、好友关系、社团驱动3个方面提取影响用户关系建立的因素并定义相应的影响因子函数。其次，针对多种影响因素难以量化以及权值分配不确定等问题，以最大熵原理为基础构建用户关系分析模型，该模型在选择特征时具有不需要依赖于特征之间的关联性等特点，并能够量化各个因素对用户关系建立的驱动强度。从而挖掘影响链接建立的关键因素，分析用户关系发展态势。实验表明，该模型不仅能够量化各因素对链接建立的驱动强度，发现关键影响因素，而且可以对用户关系进行有效预测。
- 社交网络 /
- 用户关系 /
- 关系态势 /
- 最大熵原理
Abstract: Within the evolution and development of social networks, the establishment of relationships among the users is affected by various factors. By analyzing user behavior data and relationship data in social network, this study tries to detect the key factors that affect the formation of relationship among users. Firstly, considering the complex driving factors for the user relationship establishment, the factors are extracted and the impact factor functions are defined from personal attributes, friendships and community driving. Secondly, in order to quantify driving factors and assign weight, a user relationship analysis model based on the principle of maximum entropy is proposed. The model is, when choosing features, characterized by its independence from?the association among features, and can also quantify the strength of various factors that drive users to establish relationship. Furthermore, the key factors that affect the user relationship can be detected and the development trend of user relationship can be analyzed. Experimental results reveal that the proposal model can not only quantify the strength of each factor that drives relationship establishment, it can also predict the user relationship effectively.
- Social network /
- User relationship /
- Situation analysis /
- Principle of maximum entropy

HTML全文

参考文献(21)

WANG P, XU B W, WU Y R, et al. Link prediction in social networks: The state-of-the-art[J]. Science China Information Sciences, 2015, 58(1): 1-38. doi: 10.1007/s11432-014-5237-y.

L L and ZHOU T. Link prediction in complex networks: A survey[J]. Physica A: Statistical Mechanics and Its Applications, 2011, 390(6): 1150-1170. doi: 10.1016/j.physa. 2010.11.027.

ZHU B and XIA Y. Link prediction in weighted networks: A weighted mutual information model[J]. PloS One, 2016, 11(2): e0148265. doi: 10.1371/journal.pone.0148265.

YUAN N J, ZHONG Y, ZHANG F, et al. Who will reply to/retweet this tweet: The dynamics of intimacy from online social interactions[C]. Proceedings of the Ninth ACM International Conference on Web Search and Data Mining, San Francisco, California, USA. 2016: 3-12. doi: 10.1145/ 2835776.2835800.

KAHANDA I and JENNIFER N. Using transactional information to predict link strength in online social networks[C]. International AAAI Conference on Weblogs and Social Media, California, USA, 2009: 74-81.

XIANG Rongjing, NEVILLE J, and ROGATI M. Modeling relationship strength in online social networks[C]. Proceedings of the 19th International Conference on World Wide Web, Raleigh, North Carolina, USA, 2010: 981-990. doi: 10.1145/1772690.1772790.

KASHIMA H, KATO T, YAMANISHI Y, et al. Link propagation: A fast semi-supervised learning algorithm for link prediction[C]. Proceedings of the SIAM International Conference on Data Mining, Nevada, USA, 2009, 9: 1099-1110. doi: 10.1137/1.9781611972795.94.

JONES J J, SETTLE J E, BOND R M, et al. Inferring tie strength from online directed behavior[J]. PloS One, 2013: e52168. doi: 10.1371/journal.pone.0052168.

SHAO C, DUAN Y, and WANG B. Attractive density: a new node similarity index of link prediction in complex networks[C]. IEEE International Conference on Information Science and Technology, Changsha, China, 2015: 74-78. doi: 10.1109/icist.2015.7288943.

ADAMIC L A and ADAR E. Friends and neighbors on the web[J]. Social Networks, 2003, 25(3): 211-230. doi: 10.1016/ s0378-8733(03)00009-1.

KATZ L. A new status index derived from sociometric analysis[J]. Psychometrika, 1953, 18(1): 39-43. doi: 10.1007/ bf02289026.

CHEN Y X and CHEN L. Random walks for link prediction in networks with nodes attributes[C]. Network Security and Communication Engineering: Proceedings of the 2014 International Conference on Network Security and Communication Engineering, Hong Kong, 2015: 291. doi: 10.1201/b18660-64.

XIE X, LI Y, ZHANG Z, et al. A joint link prediction method for social network[C]. International Conference of Young Computer Scientists, Engineers and Educators, Harbin, China, 2015: 56-64. doi: 10.1007/978-3-662- 46248-5_8.

LIU W and LU L. Link prediction based on local random walk[J]. Europhysics Letters, 2010, 89(5): 58007-58012. doi: 10.1209/0295-5075/89/58007.

LIU H, HU Z, HADDADI H, et al. Hidden link prediction based on node centrality and weak ties[J]. Europhysics Letters, 2013, 101(1): 18004-18009. doi: 10.1209/0295-5075/ 101/18004.

AHMED C, LKORANY A, and BAHGAT R. A supervised learning approach to link prediction in Twitter[J]. Social Network Analysis and Mining, 2016, 6(1): 1-11. doi: 10.1007/ s13278-016-0333-1.

AHEMD C and ELKORANY A. Enhancing Link prediction in twitter using Semantic user attributes[C]. Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, Paris, France, 2015: 1155-1161. doi: 10.1145/2808797.2810056.

BERGER A L, PIETRA V J D, and PIETRA S A D. A maximum entropy approach to natural language processing [J]. Computational Linguistics, 1996, 22(1): 39-71. doi: 10.1007/springerreference_179232.

PAKZAD F and ABHARI A. Characterization of user networks in Facebook[C]. Proceedings of the 2010 Spring Simulation Multiconference. Society for Computer Simulation International, Orlando, FL, USA, 2010: 105. doi: 10.1145/1878537.1878647.

PARK B, LEE K, and KANG N. The impact of influential leaders in the formation and development of social networks[C]. Proceedings of the 6th International Conference on Communities and Technologies ACM, Munich, Germany, 2013: 8-15. doi: 10.1145/2482991.2483004.

TANG X and YANG C C. Ranking user influence in healthcare social media[J]. ACM Transactions on Intelligent Systems and Technology, 2012, 3(4): 565-582. doi: 10.1145/ 2337542.2337558.

施引文献

资源附件(0)

访问统计