semantic social network analysis

24
Semantic Social Network Analysis Guillaume ERETEO

description

 

Transcript of semantic social network analysis

Page 1: semantic social network analysis

Semantic Social Network Analysis

Guillaume ERETEO

Page 2: semantic social network analysis

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]

Page 3: semantic social network analysis

Community detection

Influences the wayinformation is shared[Coleman 1988]

Influences the way actors behave[Burt 2000]

• Global structure• Distribution of actors

and activities

Page 4: semantic social network analysis

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

Page 5: semantic social network analysis

Critical mass

Page 6: semantic social network analysis

Balance Theory[Heider 1958]

Page 7: semantic social network analysis

Computer networks as social networks

[Wellman 2001]

Page 8: semantic social network analysis

web 2.0 amplifies Network effect !

Page 9: semantic social network analysis

Semantic social networks

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

Millions of FOAF profiles online

Page 10: semantic social network analysis

Social tagging

SCOT

Page 11: semantic social network analysis

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

Page 12: semantic social network analysis

The Semantic SNA Stack

Page 13: semantic social network analysis

Semantic paths in social graphs

likes

ingredient

typemainDish

Food

subclassOf

type

Page 14: semantic social network analysis

GérardGérard

FabienFabien

MylèneMylène

MichelMichelYvonneYvonne

father sister

mother

colleague

colleague

parentparentsiblingsibling

mothermotherfatherfatherbrotherbrothersistersister

colleaguecolleague

knowsknows

)( guillaumed familly

Page 15: semantic social network analysis

)( guillaumed familly

parentparentsiblingsibling

mothermotherfatherfatherbrotherbrothersistersister

colleaguecolleague

knowsknows

= 3

GérardGérard

FabienFabien

MylèneMylène

MichelMichelYvonneYvonne

father sister

mother

colleague

colleague

Page 16: semantic social network analysis

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)

Page 17: semantic social network analysis

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)

Page 18: semantic social network analysis

SemSNA an ontology of SNA

Page 19: semantic social network analysis

SemSNA an ontology of SNA

[Conein 2004][Wenger 1998]

Page 20: semantic social network analysis

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

Page 21: semantic social network analysis

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)

Page 22: semantic social network analysis

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]

Page 23: semantic social network analysis

#Guigui

#bk81

#tag27

#bk34

#tag92

#Fabien

Semantic web

Web sémantique

hasTaghasTag

hasBookmark hasBookmark

ShareInterest

MentorOf

label

label

#MichelMentorOf Collaborate

Page 24: semantic social network analysis

nameGuillaume Erétéo

organization

[email protected]

mailmentorOf

mentorOf

organizationorganization

manage

contribute

contribute answers