untitled

Size: px

Start display at page:

Download "untitled"

さわののした
7 years ago
Views:

1 SNS No Man is an Island How Relations are built within a SNS world Yuki Yasuda Yutaka Matsuo GBRC

2 The Japanese Largest Social Networking Service mixi Run by e-mercury Co.Ltd Information Exchange Relationship building Function User Network Aggregation of dyadic authentification Community Network Members Coparticipation chains communities

3 Mixi mixi Zipf (Barton196) vs

4 Barton s sigh Using random sampling of individuals, the survey is a sociological meatgrinder, tearing the individual from his social context and guaranteeing that nobody in the study interacts with anyone else in it. It is a little like a biologist putting his experimental animals through a hamburger machine and looking at every hundredth cell through a microscope; anatomy and physiology get lost, structure and function disappear, and one is left with cell biology. If our aim is to understand people s behavior rather than simply to record it, we want to know about primary groups, neighborhoods, organizations, social circles, and communities; about interaction, communication, role expectations and social control. quote from Freeman (2004 p.1)

5 Strength of Network Analysis Structural Intuition Visualization Empirical Data Mathematical/Computational Analysis With network analysis, how should we treat mixi network? What are the meaningful & important properties to look after?

6 Questions mixi mixi

7 Descriptive Statistics Data as of March 2005 # users 363,819 (Now over 1 million) # ties 381,3702 # friends (average) # communities 90,795 # community members 241,423 # community members (total) 602,8964 # members per community (average) # communities members participate # members (the largest community) # 1-member communities 5395

8 Friends Network Follows the Power Low 1 friend 85851(23.6%) Less than % Less than % Max 1301 Friends Only three members have more than 1000 friends. =2.427

9 Network of Hubs of Friends Network 99 users (0.72%) have over 300 friends. Weighted Overall Graph Clustering Coefficient=0.382 Characteristic Path Length=2.75

10 Clustering within Hubs Networks 3 bridges connect 2 cliques. # users with 200+ friends 367 (0.01%) WOGCC=0.194, L=2.88 # users with 100+ friends 2635 (0.07%) mixi

11 200

13 コミュニティIDとユーザIDの関係ユーザID平均ユーザID 最小3つの平均ユーザID最小の3つ平均ユーザID平均コミュニティID

14 Community Top10

15 Top11-20

16 Jaccard(A,B)=NA B)AUB)

17 200Jaccard0.2 Mac

18 200Jaccard0.2 Web

20 MacMax OS XPowerBook&iBook

21 Community Blockmodel Addiction ABB Appreciation Artists Jokes Hobby GLEE JAZZ TV MacipodOSX Mac User, ipod user, MacOSX user IllustratorWeb Design Web Design, Design News &

22 Density Table

23 degree rank

24 Scale free structure emerged out of strong-tie network, not weak-tie network. Scale-free structure Discovery or Creation?

25 e-mercury With Special Thanks to President Kenji, Kasahara and members of e-mercury Co.Ltd.

26 Reference A. Barton (1968)Bringing Society back in: Survey research and macro-methodology American Behavioral Scientist 12:1-9 F. Linton (2004) The Development of Social Network Analysis Empirical Press. S.P. Borgatti, M.G. Everett, and L.C. Freeman, Ucinet for Windows: Software for social network analysis (version 6.27). Harvard, Mass.: Analytic Technologies, S. Wasserman and K. Faust, Social network analysis: Methods and applications. Cambridge: Cambridge University Press. D.J. Watts and S.H. Strogatz, "Collective dynamics of "small world" networks," Nature, volume 393, pp (2005

27 SF SF SF

28 XY

untitled

untitled - - GRIPS 1 traceroute IP Autonomous System Level http://opte.org/ GRIPS 2 Network Science http://opte.org http://research.lumeta.com/ches/map http://www.caida.org/home http://www.imdb.com http://citeseer.ist.psu.edu