Helping online communities to semantically enrich folksonomies

  • View
    50

  • Download
    0

Embed Size (px)

DESCRIPTION

Helping online communities to semantically enrich folksonomies. Freddy Limpens , Fabien Gandon Edelweiss, INRIA Sophia Antipolis { freddy.limpens , fabien.gandon}@inria.fr Michel Buffa Kewi , I3S, Univerit Nice-Sophia Antipolis. Edelweiss. ISICIL, mai 2010. How to turn - PowerPoint PPT Presentation

Transcript

Helping online communities to semantically enrich folksonomies

1Helping online communities to semantically enrich folksonomiesISICIL, mai 2010Freddy Limpens, Fabien Gandon Edelweiss, INRIA Sophia Antipolis{freddy.limpens, fabien.gandon}@inria.fr

Michel BuffaKewi, I3S, Univerit Nice-Sophia Antipolis

Edelweiss

12

How to turn folksonomies ......into comprehensible topic structures ??pollutionSoil pollutionshas narrowerpollutantEnergyrelatedrelated23 without overloading users

34 and by collectingall user's expertiseinto the process

45Our approachIntegrate usage-analysis for a tailored solution

Supporting diverging points of view

Automatic processings +Human expertise through user-friendly interfaces 56Concrete scenario

Expertsproduce docs + tagArchivistscentralize + tagPublic audienceread + tag

61. Modeling statements about tags8Supporting diverging points of viewcarpollutionskos:related89carpollutionJohnPaulSupporting diverging points of viewagreesdisagreesskos:related9Supporting diverging points of view10carpollutionskos:relatedJohnPaulhasApprovedhasRejectedtagSemanticStatement (named graph)10

Supporting diverging points of view

2. folksonomy enrichment Life cycle 13DatasetDeliciousTheseNetCADICWhatBookmarks of users of tag "ademe" Keywords for Ademe's PhD projectsArchivists indexing lexicon# tags101565831439# posts101314254675# (restricted tagging)30151016025515# users8121425113ADDING TAGSAutomatic processingUser-centricstructuringDetect conflictsGlobalstructuringFlat folksonomyStructured folksonomyFolksonomy enrichment life-cycle15pollution pollutantpollutionpollutionpollutionpollutionpollutionSoil pollutions1. String-based metrics1516Evaluation of 30 edit distances

Combining the best metrics

Needs complement !1. String-based metrics16171. String-based metrics

Evaluated against Ademe expert's point of view 17

Related55,10%Broader/narrower32%CloseMatch12,82%Result on full dataset

Result on full datasetNode size InDegree tags (delicious + thesenet) mot-cls svic

Result on full datasetNode size InDegree tags (delicious + thesenet) mot-cls svic21

Fig. Markines et al. (2009)Association via :

Users

tags2. Tri-partite structure of folksonomies21tag1tag2tag3tag1freq (tag1)cooc (tag1, tag2)cooc (tag1, tag3)tag2cooc (tag2, tag1)freq (tag2)cooc (tag2, tag3)tag3cooc (tag3, tag1)cooc (tag3, tag2)freq (tag3)2. Tri-partite structure of folksonomiesTag-based association :=> gives "related" relations

Example result on CADIC dataset:2. Tri-partite structure of folksonomiesUser-based association (Mika) :environnementagriculture U1 U2 U3 U4 U6=> gives "subsumption" relations

Example result on Delicious dataset:Arrows mean "has broader"thickness weight

TheseNet dataset:Arrows mean "has broader"thickness weightADDING TAGSAutomatic processingUser-centricstructuringDetect conflictsGlobalstructuringFlat folksonomyStructured folksonomyFolksonomy enrichment life-cycle28Embedding structuring tasks within everyday activity (searching e.g)

2829Embedding structuring tasks within everyday activity (searching e.g)

2930Capturing user's point of view

3031Experimentation ADEME

31ADDING TAGSAutomatic processingUser-centricstructuringDetect conflictsGlobalstructuringFlat folksonomyStructured folksonomyFolksonomy enrichment life-cycle33Conflict detectionenvironmentpollutionnarrowerbroader3334Conflict detectionenvironmentpollutionnarrowerbroaderUsing rules e.g:

IF num(narrower)/num(broader) cTHEN narrower winsELSE 'more generic' wins3435Conflict detectionenvironmentpollutionnarrowerbroaderrelatedrelatedbroadernarrowermore genericmore generic35Ademe experimentationTotal number of relations125Contradictory43Consensual14Only rejected (to be deleted ?)2Only proposed by computer (no user review)52Debated (approved AND rejected at least once)14

Example result on Ademe's experimentation subset

Example result on Ademe's experimentation subsetADDING TAGSAutomatic processingUser-centricstructuringDetect conflictsGlobalstructuringFlat folksonomyStructured folksonomyFolksonomy enrichment life-cycle40environmentpollutionrelatedReferentUserGlobal structuring by ReferenthasApproved40ADDING TAGSAutomatic processingUser-centricstructuringDetect conflictsGlobalstructuringFlat folksonomyStructured folksonomyFolksonomy enrichment life-cycleenvironmentpollutantspollutionenvironmentpollutantspollutionnarrowernarrowerPaulenvironmentpollutionnarrowerJohnenvironmentpollutionrelatedReferentpollutantsnarrower43

Take away message (conclusion)4344Help communities

structure their tagsWhat we do :

4445Our contributions:Usages analysis

Automatic processing of tags

Tag structuring embedded in every-day tools

Supporting multi-points of view4546Future workInterfaces :to capture user-centric contributions Global administrations (Referent User)Tag searchingReal-scale experimentationMapping between knowledge representation (Gemet thesaurus Tags CADIC e.g.)Moving to other socio-structural models 4647Thank you !

freddy.limpens@inria.fr

http://www-sop.inria.fr/members/Freddy.Limpens/

http://isicil.inria.fr47484849

What is a tag ?4950Tagging model

5051Tagging model

5152HypothesesTag-link is thematicTags are concept-candidate

5253

Supporting diverging points of view5354

An eco-system of agents54555556

Tags are nice to organize your ownresources ...5657... but also to get involved in the organization of shared resources

57