基于Feeds的社交网络活跃度分析被引量：2

OSN Activities Analysis Based on User Feeds

下载PDF

导出

摘要从用户产生和消费Feeds的角度分析社交网络变得不活跃的原因,通过分析人人网某大学社区用户长周期的Feeds行为来探讨该社区用户活跃度的变化。通过对用户活跃性周期和Feeds时间间隔的分析,发现越来越多的用户产生Feeds的活跃度在下降,并导致其他用户接收到的信息流的流速和多样性下降。社交网络用户由于各种原因离开或变得不活跃,并通过信息流对其朋友圈形成负向反馈,这可能是社交网络变得不活跃的深层原因。模拟实验表明,30%的初始不活跃用户会使得整个社区的信息流快速下降,并导致整个社区不活跃。 This paper explored how one social network becomes inactive. We investigated the change of users＇ active- ness by analyzing the activeness period and Feeds inter-event time of users in a specific OSN. Our findings reveal that these users decrease their activity frequency or depart from the social network for various reasons and increase the inter- val time of Feeds, resulting in the decrease of information flow in velocity and diversity for the whole community. As a result, active users will feel the inactivity from their friends and increase the probability of being inactive as a feedback, which may be the underlying reason for the inactivity of OSN. Our simulation experiment shows that when 30% of users become inactive in generating Feeds, the whole community will be affected and collapse in a short time.

作者何旵阳孙鲁敬杨家海

机构地区汕头职业技术学院清华大学网络科学与网络空间研究院清华大学清华信息科学与技术国家实验室

出处《计算机科学》 CSCD 北大核心 2015年第11期149-153,163,共6页 Computer Science

基金国家重点基础研究发展计划(2012CB315806) 国家自然科学基金(61170211 61202356 61161140454) 博士点基金(20110002110056 20130002110058) 教育部-中国移动科研基金(MCM20123041)资助

关键词社交网络活跃度分析 Feeds行为时间间隔分析 OSN, Activities analysis, Feeds behaviors, Inter-event time analysis

分类号 TP399 [自动化与计算机技术—计算机应用技术]

引文网络
相关文献

参考文献19

1Kumar S,Zafarani R,Liu H. Understanding User MigrationPatterns in Social Media[C] // Proc. of AAAI. San Francisco,USA, Aug. 2011:1204-1209.
2Zhao X,Sala A, Wilson C, et al. Multi-scale dynamics in a mas-sive online social network[C] // Proc. of the 2012 ACM SIG-COMM conference on Internet measurement. Boston, USA,2012:171-184.
3Gong N Z,Xu W, Huang L, et al. Evolution of social-attributenetworks; measurements, modeling, and implications usinggoogle+[Cj // Proc. of the 2012 ACM conference on Internetmeasurement. Boston, USA, Nov. 2012 : 131-144.
4Leskovec J .Backstrom L.Kumar R, et al. Microscopic evolutionof social networks[C] // Proc. of the 14th ACM SIGKDD Int,lConference on Knowledge Discovery and Data Mining. Las Ve-gas, USA, Aug. 2008:462-470.
5Wu S,Das Sarma A,Fabrikant A,et al. Arrival and departuredynamics in social networks[C]//Proc. of the Sixth ACM Int’lConference on Web Search and Data Mining. Rome,Italy,Feb.2013:233-243.
6Dasgupta K,Singh R, Viswanathan B,et al. Social ties and theirrelevance to churn in mobile telecom networks[C] //Proc. of the11th intf 1 conference on Extending Database Technology: Ad-vances in database technology. Nantes, France, Mar. 2008:668-677.
7Richter Y,Yom-Tov E,Slonim N. Predicting Customer Churn inMobile Networks through Analysis of Social Groups[C] //Proc.of SDM Columbus,USA,Apr. 2010:732-741.
8Anagnostopoulos A,Kumar R,Mahdian M. Influence and corre-lation in social networks[C] //Proc. of the 14th ACM SIGKDDInt’l Conference on Knowledge Discovery and Data Mining. LasVegas,USA, Aug, 2008:7-15.
9Crandall D, Cosley D,Huttenlocher D, et al. Feedback effectsbetween similarity and social influence in online communities[G]//Proc. of the 14th ACM SIGKDD Int. 1 Conference onKnowledge EHscovery and Data Mining. Las Vegas . USA, Aug.2008:160-168.
10Backstrom L, Huttenlocher D,Kleinberg J,et al. Group forma-tion in large social networks:membership,growth,and evolution[C] // Proc. of the 12th ACM SIGKDD Int.l Conference onKnowledge discovery and data mining. Philadelphia,USA,Aug.2006:44-54.