基于隐回归的用户关系强度模型

doi:10.11772/j.issn.1001-9081.2016.02.0336

计算机应用 ›› 2016, Vol. 36 ›› Issue (2): 336-341.DOI: 10.11772/j.issn.1001-9081.2016.02.0336

• 第三届CCF大数据学术会议(CCF BigData 2015) • 上一篇下一篇

基于隐回归的用户关系强度模型

韩忠明, 谭旭升, 陈炎, 杨伟杰

北京工商大学计算机与信息工程学院, 北京 100048

收稿日期:2015-08-29 修回日期:2015-09-13 出版日期:2016-02-10 发布日期:2016-02-03
通讯作者: 韩忠明(1972-),男,山西文水人,副教授,博士,CCF会员,主要研究方向:大数据处理、数据挖掘。
作者简介:谭旭升(1989-),男,广西柳州人,硕士研究生,主要研究方向:数据挖掘;陈炎(1991-),男,江苏泰州人,硕士研究生,主要研究方向:数据挖掘;杨伟杰(1980-),女,山东潍坊人,讲师,博士,CCF会员,主要研究方向:数据挖掘。
基金资助:
国家自然科学基金资助项目(61170112);教育部人文社会科学研究基金资助项目(13YJC860006);北京市属高等学校科学技术与研究生教育创新工程建设项目(PXM2012_014213_000037)。

Strength model of user relationship based on latent regression

HAN Zhongming, TAN Xusheng, CHEN Yan, YANG Weijie

School of Computer and Information Engineering, Beijing Technology and Business University, Beijing 100048, China

Received:2015-08-29 Revised:2015-09-13 Online:2016-02-10 Published:2016-02-03

摘要/Abstract

摘要： 为了科学合理地度量社会网络中用户间的有向关系强度,基于用户有向交互次,提出一个度量用户交互强度的光滑模型。将用户关系强度作为隐变量,交互强度作为因变量,构建度量用户关系强度的隐变量回归模型,并给出求解隐变量回归模型的最大期望(EM)算法。分别从人人网和新浪微博采集了数据集,从最佳好友、强度排序等方面进行了大量实验。在人人网实验中,通过关系模型选择的TOP-10好友与人工标注结果比较,得出NDCG均值为69.48%,平均查准率均值(MAP)为66.3%,与对比算法相比有明显提高;在大规模新浪微博数据集实验中,将关系强度大的节点作为传染模型的源节点的传播范围相较于选择其他节点作为源节点平均提高了80%。实验结果说明所提模型能够有效度量用户间的关系强度。

关键词: 关系强度, 社会网络, 隐变量回归模型

Abstract: To effectively measure the strength of the directed relationship among the users in social network, based on the directed interaction frequency, a smooth model for computing the interaction strength of the user was proposed. Furthermore, user interaction strength was taken as dependent variable and user relationship strength was taken as latent variable, a latent regression model was constructed, and an Expectation-Maximization (EM) algorithm for parameter estimation of the latent regression model was given. Comprehensive experiments were conducted on two datasets extracted from Renren and Sina Weibo in the aspects of the best friends and the intensity ranking. On Renren dataset, the result of TOP-10 best friends chosen by the proposed model was compared with that of manual annotation, the mean of Normalized Discounted Cumulative Gain (NDCG) of the model was 69.48%, the average of Mean Average Precision (MAP) of the model was 66.3%, both of the parameters were significantly improved; on Sina Weibo dataset, the range of infection spread by nodes with higher relationship strength increased by 80% compared to the other nodes. The experimental results show that the proposed model can effectively measure user relationship strength.

Key words: relationship strength, social network, latent regression model

中图分类号:

TP181

韩忠明, 谭旭升, 陈炎, 杨伟杰. 基于隐回归的用户关系强度模型[J]. 计算机应用, 2016, 36(2): 336-341.

HAN Zhongming, TAN Xusheng, CHEN Yan, YANG Weijie. Strength model of user relationship based on latent regression[J]. Journal of Computer Applications, 2016, 36(2): 336-341.

参考文献

[1] GRANOVETTER M S. The strength of weak ties[J]. American Journal of Sociology, 1973, 78(2): 1360-1380.
[2] ZHAO X, YUAN J, LI G, et al. Relationship strength estimation for online social networks with the study on Facebook[J]. Neurocomputing, 2012, 95(14): 89-97.
[3] KAHANDA I, NEVILLE J. Using transactional information to predict link strength in online social networks[C]//Proceedings of the Third International Conference on Weblogs and Social Media. Menlo Park, CA: AAAI Press, 2009: 74-81.
[4] GILBERT E, KARAHALIOS K. Predicting tie strength with social media[C]//CHI '09: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. New York: ACM, 2009: 211-220.
[5] HAYTHORNTHWAITE C. The strength and the impact of new media[C]//HICSS '01: Proceedings of the 34th Annual Hawaii International Conference on System Sciences. Washington, DC: IEEE Computer Society, 2001, 1: 1019.
[6] GILSING V, NOOTEBOOM B. Density and strength of ties in innovation networks: an analysis of multimedia and biotechnology[J]. European Management Review, 2005, 2(3): 179-197.
[7] WHITE D R, HOUSEMAN M. The navigability of strong ties: Small worlds, tie strength, and network topology[J]. Complexity, 2002, 8(1): 72-81.
[8] JONES J J, SETTLE J E, BOND R M, et al. Inferring tie strength from online directed behavior[J]. Plos One, 2013, 8(1): e52168-e52168.
[9] SARAMÄKI J, LEICHT E A, LÓPEZ E, et al. Persistence of social signatures in human communication[J]. Proceedings of the National Academy of Sciences, 2014, 111(3): 942-947.
[10] XIANG R, NEVILLE J, ROGATI M. Modeling relationship strength in online social networks[C]//WWW '10: Proceedings of the 19th International Conference on World Wide Web. New York: ACM, 2010: 981-990.
[11] SINGLA P, RICHARDSON M. Yes, there is a correlation——from social networks to personal behavior on the Web[C]//WWW 2008: Proceedings of the 17th International Conference on World Wide Web. New York: ACM, 2008: 655-664.
[12] MOUSTAKI I, KNOTT M. Generalized latent trait models[J]. Psychometrika, 2000, 65(3): 391-411.
[13] ONNELA J P, SARAMÄKI J, HYVÖNEN J, et al. Structure and tie strengths in mobile communication networks[J]. Proceedings of the National Academy of Sciences, 2007, 104(18): 7332-7336.
[14] ZHANG H, DANTU R. Predicting social ties in mobile phone networks[C]//Proceedings of the 2010 IEEE International Conference on Intelligence and Security Informatics. Piscataway, NJ: IEEE, 2010: 25-30.
[15] HIDALGO C A, RODRIGUEZ-SICKERT C. The dynamics of a mobile phone network[J]. Physica A: Statistical Mechanics and Its Applications, 2008, 387(12): 3017-3024.
[16] BAYM N K, LEDBETTER A. Tunes that bind? Predicting friendship strength in a music-based social network[J]. Information, Communication & Society, 2009, 12(3): 408-427.
[17] DE CHOUDHURY M, MASON W A, HOFMAN J M, et al. Inferring relevant social networks from interpersonal communication[C]//WWW '10: Proceedings of the 19th International Conference on World Wide Web. New York: ACM, 2010: 301-310.
[18] ZHUANG J, MEI T, HOI S C H, et al. Modeling social strength in social media community via kernel-based learning[C]//MM '11: Proceedings of the 19th ACM International Conference on Multimedia. New York: ACM, 2011: 113-122.
[19] HU B, JAMALI M, ESTER M. Learning the strength of the factors influencing user behavior in online social networks[C]//ASONAM '12: Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining. Washington, DC: IEEE Computer Society, 2012: 368-375.
[20] XLIN X, SHANG T, LIU J. An estimation method for relationship strength in weighted social network graphs[J]. Journal of Computer and Communications, 2014, 2(4): 82-89.
[21] SPILIOTOPOULOS T, PEREIRA D, OAKLEY I. Predicting tie strength with the Facebook API[C]//PCI '14: Proceedings of the 18th Panhellenic Conference on Informatics. New York: ACM, 2014: 1-5.
[22] GILBERT E. Predicting tie strength in a new medium[C]//CSCW '12: Proceedings of the ACM 2012 Conference on Computer Supported Cooperative Work. New York: ACM, 2012: 1047-1056.
[23] SERVIA-RODRÍGUEZ S, DÍAZ-REDONDO R P, FERNÁNDEZ-VILAS A, et al. A tie strength based model to socially-enhance applications and its enabling implementation: mySocialSphere[J]. Expert Systems with Applications, 2014, 41(5): 2582-2594.
[24] WANG B K, MIN Y L, HUANG Y F, et al. Review rating prediction based on the content and weighting strong social relation of reviewers[C]//UnstructureNLP '13: Proceedings of the 2013 International Workshop on Mining Unstructured Big Data Using Natural Language Processing. New York: ACM, 2013: 23-30.
[25] KIM J, LEE E, CHOI J, et al. Monitoring social relationship among Twitter users by using NodeXL[C]//Proceedings of the 2013 Research in Adaptive and Convergent Systems. New York: ACM, 2013: 107-110.
[26] TARPEY T, PETKOVA E. Latent regression analysis[J]. Statistical Modelling, 2010, 10(2): 133-158.
[27] JÄRVELIN K, KEKÄLÄINEN J. Cumulated gain-based evaluation of IR techniques[J]. ACM Transactions on Information Systems, 2002, 20(4): 422-446.
[28] MADRIA S K, BHOWMICK S S, NG W K, et al. Research issues in Web data mining[C]//DaWaK '99: Proceedings of the First International Conference on Data Warehousing and Knowledge Discovery, LNCS 1676. Berlin: Springer, 1999: 303-312.

基于隐回归的用户关系强度模型

Strength model of user relationship based on latent regression

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	孙鹤立, 孙玉柱, 张晓云. 基于生成对抗网络的事件描述生成[J]. 计算机应用, 2021, 41(5): 1256-1261.
[2]	孙鹤立, 孙玉柱, 张晓云. 基于事件描述的社交事件参与度预测[J]. 计算机应用, 2020, 40(11): 3101-3106.
[3]	秦海燕, 章永龙, 李斌. 社会网络下分配众包任务的真实机制[J]. 计算机应用, 2020, 40(10): 3019-3024.
[4]	孙子力, 彭舰, 仝博. 社会网络中基于社群衰减的影响力最大化算法[J]. 计算机应用, 2019, 39(3): 834-838.
[5]	郑晓雪, 张大方, 刁祖龙. 基于用户身份特征的多标签分类算法[J]. 计算机应用, 2017, 37(6): 1697-1701.
[6]	刘院英, 郭景峰, 魏立东, 胡心专. 成本控制下的快速影响最大化算法[J]. 计算机应用, 2017, 37(2): 367-372.
[7]	李志明, 唐永中. 基于社交网络和关联数据的服务网络构建方法[J]. 计算机应用, 2017, 37(2): 468-472.
[8]	王珊珊, 冷甦鹏. 面向移动社会网络的好友推荐方法[J]. 计算机应用, 2016, 36(9): 2386-2389.
[9]	徐涛, 孟野. 踪迹聚类下组织实体的重要度排序方法[J]. 计算机应用, 2016, 36(5): 1284-1289.
[10]	李源昊, 陆平, 吴一凡, 韦薇, 宋国杰. 面向移动社会网络的用户年龄与性别特征识别[J]. 计算机应用, 2016, 36(2): 364-371.
[11]	张晓琳, 王萍, 郭彦磊, 王静宇. 社会网络子集(θ, k)-匿名方法[J]. 计算机应用, 2015, 35(8): 2178-2183.
[12]	孙辉霞, 李跃新. 基于潜在特征的重叠社团识别算法[J]. 计算机应用, 2015, 35(12): 3477-3480.
[13]	李美子, 向阳, 张波, 金波. 社会网络中基于信任链的主题群组发现算法[J]. 计算机应用, 2015, 35(1): 157-161.
[14]	夏涛陈云芳张伟陆有为. 社会网络中的影响力综述[J]. 计算机应用, 2014, 34(4): 980-985.
[15]	李慧胡云施珺. 社会网络环境下的协同推荐方法[J]. 计算机应用, 2013, 33(11): 3067-3070.