on Social Network Analysis

steelsquareInternet και Εφαρμογές Web

20 Οκτ 2013 (πριν από 3 χρόνια και 1 μήνα)

65 εμφανίσεις

A State of the Art

on Social Network Analysis
and its Applications on

a Semantic Web

Guillaume ERETEO, Michel Buffa,
Fabien Gandon, Patrick Grohan,
Mylène Leitzelman, Peter Sander

What's going on?


Social network analysis (SNA) :


Indices, Algorithms



Social network analysis and the web


Web "1.0"


Web 2.0


Semantic web



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 communication


To improve network resilience


To trust

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

Community

detection

Influences the way


information is shared

[Coleman 1988]
Influences the way actors behave

[Burt 2000]



Global structure



Distribution of actors
and activities

Community detection


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]

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]

Betweenness Centrality


Exact computation O(n.m)


[Newman 2001] [Brandes 2001]



Approximation :


[Brandes et al 2007] [Bader et al 2007]
[Geisberg et al 2008]



Parallel algorithms


[Bader et al 2006] [Santos et al 2006]


Computer networks as social networks



Asynchonous and synchronous
conversation : mail, IRC, chat


[Tyler et al 2003]


Hyperlink structure of homepages


[Adamic and Adar 2003]


Cooccurence of names in web pages


[Kautz el al 1997][Mika 2005]


[Matsuo et al 2006][Jin et al 2007]



[Wellman 2001]


Network effect amplified by web 2.0

Semantic social networks






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


Millions of FOAF profiles
online

SNA on the semantic web



Rich graph representations are always reduced
to simple untyped graphs in order to apply SNA



[Golbeck et al 2003] [Golbeck and
Rothstein

2008]

[Paolillo and Wright 2006]

Foaf:knows

Foaf:interest

Semantic SNA



Ndegree with SPARQL and the

semantic engine CORESE

construct
{


?y
semsna:hasInDegree

_:bO



_:bO
semsna:isDefinedForProperty

rel:worksWith



_:bO
semsna:hasValue

?indegree


_:b0 semsna:hasDistance 2

}

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


?x
$path

?y


filter(match($path,
star(foaf:knows)
))


fitler(
pathLength($path)<= 2
)

}group by ?y


filter(match($path,
star(
rel:worksWith
)
))

Select the most popular actors in a
work subnetworks


select ?x ?indegree{


?x
semsna:hasInDegree

?z


?z
semsna:isDefinedForProperty rel:worksWith


?z
semsna:hasValue

?indegree


?z
semsna:hasDistance

2

}

order by desc(?indegree)



Pre
-
process betweenness centrality


PREFIX foaf: <http://xmlns.com/foaf/0.1/>

select ?from ?to ?between $path
pathLength($path)

as ?length
where{


?from

$path

?to


graph $path{?between foaf:knows ?j}


filter(match($path, star(foaf:knows)))


filter(?from != ?to)


filter(?from != ?between)


filter(?between != ?to)


optional { ?from foaf:knows::?p ?to }


filter(!bound(?p))

}

group by $path

order by ?length

detect paths between all pairs of indirectly connected
resources and their intermediaries

Questions

?

Thank you for your attention