• Like
  • Save
Social Networks Analysis: challenges in the era of the social web
Upcoming SlideShare
Loading in...5
×
 

Social Networks Analysis: challenges in the era of the social web

on

  • 1,182 views

Conférence thématique à la journée futur & ruptures de l'Institut Telecom

Conférence thématique à la journée futur & ruptures de l'Institut Telecom
http://www.fondation-telecom.org/evenement/journee-futur-et-ruptures-37/

Statistics

Views

Total Views
1,182
Views on SlideShare
1,105
Embed Views
77

Actions

Likes
0
Downloads
16
Comments
0

3 Embeds 77

http://labocommunicant.net 74
http://www.linkedin.com 2
https://twitter.com 1

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Social Networks Analysis: challenges in the era of the social web Social Networks Analysis: challenges in the era of the social web Presentation Transcript

    • Social Network AnalysisChallengesCécile Bothorel
    • Social Network Analysis  Userspoint of view  Analyst point of view2 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • An old field of Social Sciences 1933 : sociogram by Jacob Levy Moreno to map relationships 1954 : John A. Barnes introduces "social network" 1967 : Six degrees of separation by Stanley Milgram 1973 : "The Strength of Weak Ties" by Mark L. Granovetter http://fr.wikipedia.org/wiki/Fichier:Six_degrees_of_separation.png3 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • A new umbrella term: Complex Networks  Macroscopic structure of the web, late 1990s  Real data in a wide • 2 Altavista crawls each with variety of sources over 200 million pages and 1.5 billion links • Biology, physics, finance, medecine, transportation, computer sciences  Whats in common? Understand • General properties, structure, evolution, propagation4 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • (very) big, rich and dynamic graphs5 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • (very) big, rich and dynamic graphs Now online social media available A new computational social science Challenges at every steps • Data management • Model design • Analysis : measures, mining algorithms, propagation, evolution • Visualisation6 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Data management  Sources  Acquisition  Preparation  Storage  « Big data », cloud,  Update parellelization  Privacy: anonymization not enough? [Backstrom et al. 2007]  Data owners do not know how to exploit... and manage7 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Model  Explicit/implicit connections  Types of relationship  Details interactions  Profiles  …..8 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Analysis - measures  Lots of measures  But which ones? Prestige Influence Centrality  Marketing KPI: Relevance PageRank • Value of Rts in Twitter? • Value of a fan Facebook?  Expert-finding  E-Reputation  Recommendation, social Shopping9 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Analysis – community detection [Fortunato 2010, Pons 2007]  Hierarchical algorithms  Modularity maximisation  Spectral partitioning  Evaluation?  What is a good partition?10 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Analysis – community detection Structural partition only What about Augmented Social network? • Additional information: Users profile, shared contents, opinions, etc. Network Dynamics? • Graph evolution [Leskovec et al. 2005, 2008] • Link prediction [Liben- Nowell and Kleinberg 2003] [Cruz et al. 2010, 2011 Thèse Futur&Ruptures]11 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Analysis – propagation, virality  Influence - information propagation  Maximizing the spread of influence  Influence may come from outside the monitored data  Part of influence-based contagion and homophily-driven [Easley and Kleinberg 2010] diffusion? Which factors affects [Leskovec et al. 2006, 2007] the propagation ?12 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Visualisation – for the end-user13 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Visualisation – for the analyst  Open Tools: Gephi, Tulip, iGraph  Plenty of measures, algorithms  Volume of Data? Artists in the last.fm database, by Tamas Nepusz14 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Visualisation – for the analyst  New layouts?  Here: Analyse the interaction between communities [Cruz et al. 2012, Thèse Futur&Ruptures]15 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Visualisation – for the analyst  Augmented social network exploration • Compare perspectives • Differents role assignation [Cruz et al. 2012, Thèse Futur&Ruptures]16 Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012
    • Conclusion Social network analysis: a renewed research area... … to be challenged. Various open questions & difficulties. With sample of our contribution: Augmented Social Network Analysis, Futur&Ruptures PhD thesis, Institut Telecomhttp://http://labocommunicant.net/ppc/cecile-bothorel/http://http://perso.telecom-bretagne.eu/cecilebothorel/http://twitter.com/#!/CecileBothorel [F. Bonchi et al. 2011, Social network analysis and Mining for Business applications, ACM Trans. Intell. Syst. Technol.] Telecom Bretagne, LUSSI Journées Futur & Ruptures, 26 January 2012