semantic social network analysis

Semantic Social Network Analysis

Guillaume ERETEO

Social Network Analysis?

• A science to understand the structure, the interactions and the strategic positions in social networks.

• Sociograms[Moreno, 1933]

• What for? – To control information flow– To improve/stimulate communication– To improve network resilience– To trust

[Wasserman & Faust 1994] [Scott 2000] [Mika 2007]

Community detection

Influences the wayinformation is shared[Coleman 1988]

Influences the way actors behave[Burt 2000]

• Global structure• Distribution of actors

and activities

Centrality: strategic positions

Degree centrality: Local attention

beetweenness centrality:reveal broker "A place for good ideas"[Burt 1992] [Burt 2004]

Closeness centrality: Capacity to communicate

[Freeman 1979]

Community detection: Distribution of actors and activities

Critical mass

Balance Theory[Heider 1958]

Computer networks as social networks

[Wellman 2001]

web 2.0 amplifies Network effect !

Semantic social networks

http://sioc-project.org/node/158

Millions of FOAF profiles online

Social tagging

SCOT

SNA on the semantic web

Rich graph representations reduced to simpleuntyped graphs in order to apply SNA

[Paolillo and Wright 2006]

Foaf:knows

Foaf:interest

The Semantic SNA Stack

Semantic paths in social graphs

likes

ingredient

typemainDish

Food

subclassOf

type

GérardGérard

FabienFabien

MylèneMylène

MichelMichelYvonneYvonne

father sister

mother

colleague

colleague

parentparentsiblingsibling

mothermotherfatherfatherbrotherbrothersistersister

colleaguecolleague

knowsknows

)( guillaumed familly

)( guillaumed familly

parentparentsiblingsibling

mothermotherfatherfatherbrotherbrothersistersister

colleaguecolleague

knowsknows

= 3

GérardGérard

FabienFabien

MylèneMylène

MichelMichelYvonneYvonne

father sister

mother

colleague

colleague

select ?y ?to pathLength($path) as ?length sum(?length) as ?centrality where{

?y $path ?tofilter(match($path, star(param[type]param[type]), 'sa'))

}group by ?y

Closeness centrality

Cc<type>(y)

add{?x semsna:isMemberOf ?uri

}select ?x ?y genURI(<myorg>) as ?uri from Gwhere { ?x $path ?y filter(match($path, star(param[type]param[type]), 'sa'))}group by any

Parametrized ComponentC<type>(G)

SemSNA an ontology of SNA

SemSNA an ontology of SNA

[Conein 2004][Wenger 1998]

Parametrized n-Degree

construcconstructt{?y semsna:hasInDegreesemsna:hasInDegree _:bO _:bO semsna:isDefinedForPropertysemsna:isDefinedForProperty param[type] _:bO semsna:hasValuesemsna:hasValue ?indegree_:b0 semsna:hasDistance param[length]param[length]

}select ?y count(?x) as ?indegree{

?x $path$path ?y filter(match($path, star(star(param[type]param[type]))))fitler(pathLength($path) <= pathLength($path) <= param[length]param[length])

}group by ?y

Most popular manager in a work subnetworks

select ?y ?indegree{

?y rdf:type domain:Manager

?y semsna:hasInDegreesemsna:hasInDegree ?z

?z semsna:isDefinedForProperty semsna:isDefinedForProperty rel:worksWithrel:worksWith

?z semsna:hasValuesemsna:hasValue ?indegree

?z semsna:hasDistancesemsna:hasDistance 2

}

order by desc(?indegree)

Current Community detection algorithms

• Hierarchical algorithms

– Agglomerative (based on vertex proximity):• [Donetti and Munoz 2004] [Zhou Lipowsky, R. 2004]

– Divisive (mostly based on centrality):• [Girvan and Newman 2002] [Radicchi et al 2004]

• Based on heuristic (modularity, randon walk, etc.)

• [Newman 2004], [Pons and Latapy 2005], [Wu and Huberman 2004]

#Guigui

#bk81

#tag27

#bk34

#tag92

#Fabien

Semantic web

Web sémantique

hasTaghasTag

hasBookmark hasBookmark

ShareInterest

MentorOf

label

label

#MichelMentorOf Collaborate

nameGuillaume Erétéo

organization

[email protected]

mailmentorOf

mentorOf

organizationorganization

manage

contribute

contribute answers