www-tutorial-close-final [Compatibility  Mode]
5 Pages
English
Downloading requires you to have access to the YouScribe library
Learn all about the services we offer

www-tutorial-close-final [Compatibility Mode]

-

Downloading requires you to have access to the YouScribe library
Learn all about the services we offer
5 Pages
English

Description

™™‰‰™‰‰™‰‰‰Detecting, Understanding and Exploiting Web communitiesClosureby Athena Vakali & Ioannis KompatsiarisConclusionsin summary ...different community definitions – most of them rely on structure – Web relevant : graph/spectral; overlapping & interactions-driven (Web 2.0) significant number of community detection methodologies - Web relevant : centralityqy/cliques, Bippgartite graph-based,y, Modularity optimization (in general: computationally efficient approaches)Web graph visualization - Web relevant : communities are not always “guaranteed”exploitations of communities on the Web :navigation, retrievalrecommendationssocial Web 2.0content outsourcing....Detecting, Understanding and Exploiting Web communitiesAthena Vakali & Ioannis Kompatsiaris 2™™™™™™™™™™™Trends Trends - - IIssues ssues (()I)huge scales on the Web – communities to support compression – algorithms optimizationexitistiing experimentattiions on fftragmented datasets -standard test sets for comparisons are neededprohibiting complexities require approximationscommunities evaluation – more than the modularity measureDetecting, Understanding and Exploiting Web communitiesAthena Vakali & Ioannis Kompatsiaris 3Trends -Trends - IIssues ssues (()II) communities evolution & dynamics (time, social/semantic criteria)embed communities to the informationinformation rretrievaletrieval processcommunities interpretation … are they meaningful/useful ...

Subjects

Informations

Published by
Reads 8
Language English

Exrait

Detecting Understanding and
Detecting Understanding and
Detecting, Understanding and
Detecting, Understanding and
Exploiting Web communities
Exploiting Web communities
Closure
Closure
by Athena Vakali & Ioannis Kompatsiaris
Conclusions
Conclusions
in summary ...
™
different community definitions – most of them rely on structure –
Web
relevant :
graph/spectral; overlapping & interactions-driven (Web 2.0)
™
significant number of community detection methodologies
-
Web relevant :
centrality/cliques, Bipartite graph-based, Modularity optimization (in
general: computationally efficient approaches)
™
Web graph visualization -
Web relevant :
communities are not always
“guaranteed”
™
exploitations of communities on the Web :
‰
navigation, retrieval
‰
recommendations
‰
recommendations
‰
social Web 2.0
‰
content outsourcing
‰
....
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
2
Trends
Trends - Issues (I)
Issues (I)
™
huge
scales
on the Web – communities to support
compression
algorithms optimization
compression – algorithms optimization
™
existing experimentations on
fragmented datasets
-
standard test sets for comparisons are needed
™
prohibiting
complexities
require approximations
™
communities
evaluation
– more than the modularity
measure
measure
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
3
Trends
Trends - Issues (II)
Issues (II)
™
communities
evolution & dynamics
(time,
social/semantic criteria)
™
embed communities
to the information retrieval
™
embed communities
to the information retrieval
process
™
communities
interpretation
… are they
meaningful/useful ?
™
communities for
tagging
networks
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
4
some indicative work for
some indicative work for
inspiring future research efforts
inspiring future research efforts
™
Incorporating community models in search
engines
™
Communities as a barometer of the
blogosphere
blogosphere
™
Enriching semantic web ontologies with
community models
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
5
Incorporating community
models in search engines
Exploitation of web user sessions
Æ
Session Interest
Graphs
Æ
Community detection & summarization
Æ
Graphs
Æ
Community detection & summarization
Æ
Assignment of new sessions to one of existing
communities
Æ
Use of Bayesian models for incorporating
community information in retrieval
[Almeida04]
Conventional retrieval model
Community-aware retrieval model
Conventional retrieval model
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
6
Communities as a barometer
of the blogosphere
of the blogosphere
™
Creation of the blog
™
Creation of the blog
graphs
Æ
community
detection
Æ
tensor
politics
model
Æ
community
evolution
[Chi2007]
hurricane Katrina
community temporal trends
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
7
Enriching SW ontologies with
community models
community models
Exploitation of user interactions online (particularly in Web 2.0 applications)
Æ
Graph representation
Æ
Community detection
Æ
Incorporation of
Graph representation
Æ
Community detection
Æ
Incorporation of
community models in the ontology engineering process
[Specia07], [Mika05]
The extracted ontologies will:
‰
Better reflect the knowledge model as perceived by each community
‰
Embed temporal dimension
ii) Localized ontologies per user community
i) Group of relevant concepts emerged from del.icio.us
User
Community
User
Community
Mapping
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
8
Community
References
™
[Almeida04] Almeida, R. B. and Almeida, V. A. 2004. A community-aware search
i
I
P
di
f th 13th i t
ti
l C
f
W ld Wid W b
(N
engine. In
Proceedings of the 13th international Conference on World Wide Web
(New
York, NY, USA, May 17 - 20, 2004). WWW '04. ACM, New York, NY, 413-421.
™
[Chi07] Chi, Y., Zhu, S., Song, X., Tatemura, J., and Tseng, B. L. 2007. Structural and
temporal analysis of the blogosphere through community factorization. In
Proceedings
of the 13th ACM SIGKDD international Conference on Knowledge Discovery and Data
of the 13th ACM SIGKDD international Conference on Knowledge Discovery and Data
Mining
(San Jose, California, USA, August 12 - 15, 2007). KDD '07. ACM, New York,
NY, 163-172.
™
[Specia07] Specia, L. and Motta, E. 2007. Integrating Folksonomies with the Semantic
Web. In Proceedings of the 4
th
European Semantic Web Conference (ESWC ’07),
Innsbruck, Austria, pp. 624-639
™
[Mika05] Mika, P. 2005. Ontologies Are Us: A Unified Model of Social Networks and
Semantics. In Proceedings of
the 4
th
International Semantic Web Conference (ISWC’
05), Galway, Ireland, pp. 522-536
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
9
Athena Vakali & Ioannis Kompatsiaris
Detecting, Understanding and Exploiting Web communities
10