; Link Prediction in Multi Modal Social Networks
Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Link Prediction in Multi Modal Social Networks


  • pg 1
									 In the name of God

Course Title:
Social Networks

Instructor: Masoud      Asadpour
Lecture: Introduction
The Wisdom of Crowds
n    (Surowiecki,2004), these key criteria separate
      wise crowds from irrational ones:
      ¨  Diversity of opinion: Each person should have private
          information even if it's just an eccentric interpretation of
          the known facts.
      ¨  Independence: People's opinions aren't determined by
          the opinions of those around them.
      ¨  Decentralization: People are able to specialize and
          draw on local knowledge.
      ¨  Aggregation: Some mechanism exists for turning
          private judgments into a collective decision.
Examples of Networks
n    Brain interconnectivity map
n    Ecosystem and food chains
n    Gene, protein, molecule, and virus networks
n    Political relations and Lobbies
n    Companies and their financial relations
n    Bank transactions
n    Power lines, phone lines
n    Roads, highways, airlines, railways
n    Paper citations
n    Emails, Web, Internet
n    Mobile, SMS and Phone networks
n    Online friendship and social networks
A friendship network on
     Friendship network of children in
     a US school
                                      Yellow - White Race
                                      Green - Black Race
                                      Pink - Other

    Hierarchical structure of the

     interconnected cities across the
     globe (by router configuration and not physical backbone)

Iranian Censored Network
Influence Netwotk in
Poles in Balatarin.com
Road and Airlines Network

Genetic interaction network
Yeast protein-protein interaction
    Food Web

Related courses
n  SocialNetworks
n  Complex Networks
n  Dynamic Networks
n  Prerequisites
   ¨ Algorithms
   ¨ Discrete
   ¨ Graph Theory
Examples of Applications
n  Elections
n  Marketing (e.g. viral marketing)
n  Revolutions
n  Epidemics
n  Fraud detection
n  ...?
What makes Social Networks
n  Astring of recent discoveries has forced us
  to acknowledge that amazingly simple and
  far-reaching natural laws govern the
  structure and evolution of all the complex
  networks that surround us
Why Network Science is important?
n  Reductionism:    to disassemble nature in
    order to understand it
n  After spending a lot to disassemble nature,
    now we almost know the pieces. But we are
    still far from understanding nature as a
n  Today we recognize that nothing happens
    in isolation. Most events are inter-
Why Network Science is important?
n  Nowadays, power lies within the hands of
   the ones who know the network:
    ¨ Brokering
    ¨ Public relation
    ¨ Advertisement
    ¨ Communication
    ¨ …?

n  Inthis course you –hopefully- learn how to
   think network!
n    Easley, Kleinberg, Networks, Crowds, and Markets
      Reasoning about a Highly Connected World, 2010
n    Barabasi, Linked the new science of networks, 2002
n    S. Wasserman and K. Faust, Social Network Analysis,
n    More readings:
      ¨  P.J. Carrington, J. Scott, S. Wasserman, models and methods in
          social network analysis, 2005
      ¨  Hanneman, Introduction To Social Networks Methods, 2005
      ¨  J. Scott, Social Network Analysis: A Handbook, 2000
      ¨  A. Degenne and M. Forse, Introducing Social Networks, 1999
      ¨  Books on Graph Theory, Random Graphs, …
n    Boccaletti et al, “Complex networks Structure
      and Dynamics,” 2006
n    Newman, “The structure and function of complex
      networks,” 2003
n    Costa et al, “Characterization of complex
      networks A survey of measurements,” 2008
n    Fortunato, Community detection in graphs, 2010
n    Arenas et al, “synchronization in complex
      networks,” 2008
Grade (subject to change)
n  40% Homework
n  30% final exam
n  30% final project

n  Slideswill be available on ece.ut.ac.ir/
n  Register in CECM to receive the HWs
Software you might work with:
n    Network Analysis Tool:
      ¨    Network Workbench
      ¨    Pajek
      ¨    UCI Net
      ¨    NetworkX
      ¨    Gephi
n    Graph library:
      ¨    Jung
      ¨    SNAP library (C++): large graphs
      ¨    BOOST graph library (C++): uses STL
n    Data mining:
      ¨    Clementine
Lectures (perhaps not in this order)
n    Random graphs, power law, small-world property, scale-
      free networks, and generative processes
n    Networks and measurements: centrality and prestige,
      degree, closeness, betweenness, information, rank
      centralities, etc
n    Structural balance, transitivity, clusterability, motifs,
      dyadic and triadic relations
n    Communities and cohesive subgroups, clique, n-cliques,
      k-plexes, k-cores
n    Structural equivalence, positions, roles, and blockmodels
n    Diffusion of Information, Failures and Epidemics in
n    Consensus, Synchronization, etc
Structural Equivalence
             H                       G               B
     G               B
             A                       F               C
    F                C
                                         E       D
         E       D

             A   B       C   D   E       F       G
List of projects (Groups of 2 or 3

n  Social mobile games
n  Social networking on mobile platforms
n  Bluetooth networks
n  Facebook applications
n  Social networking plugin for FireFox
List of projects
n  Social Network Engine
n  Social Network Analysis Toolbox
n  Search Engine based on Social Networks
n  Search Engine Optimization
n  Analysis of a subset of Persian web
   ¨ Persian   blogsphere
   ¨ Political groups

n  Persian   WordNet, semantic web
List of projects
n  Web   content analysis
  ¨    Keyword extraction
  ¨    Blog post extraction
  ¨    Expertness measuring
  ¨    Topic Modeling
  ¨    Diffusion of information and news
  ¨    Trend analysis
  ¨    Impact of online contents on real world
List of projects
n    Simulation of social movements
n    Control of collective behavior
n    Simulation of financial transactions
      ¨    Anti-money laundering
      ¨    Fraud detection
n    Network Marketing
n    Socially spreading diseases
n    Collection or Analysis of real world social networks
      ¨    NA
      ¨    Handicapped people
n    You can suggest project…

To top