57
1 Helping online communities to semantically enrich folksonomies ISICIL, mai 2010 Freddy Limpens, Fabien Gandon Edelweiss, INRIA Sophia Antipolis {freddy.limpens, fabien.gandon}@inria.fr Michel Buffa Kewi, I3S, Univerité Nice-Sophia Antipolis Edelweiss

Helping online communities to semantically enrich folksonomies

  • Upload
    tavon

  • View
    59

  • 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, Univerité Nice-Sophia Antipolis. Edelweiss. ISICIL, mai 2010. How to turn - PowerPoint PPT Presentation

Citation preview

Page 1: Helping online communities to semantically enrich folksonomies

1

Helping online communities to semantically enrich folksonomies

ISICIL, mai 2010

Freddy Limpens, Fabien Gandon Edelweiss, INRIA Sophia Antipolis

{freddy.limpens, fabien.gandon}@inria.fr

Michel BuffaKewi, I3S, Univerité Nice-Sophia Antipolis

Edelweiss

Page 2: Helping online communities to semantically enrich folksonomies

2

How to turn folksonomies ...

...into comprehensible topic structures ?

?pollution

Soil pollutions

has narrower

pollutant Energy

related related

Page 3: Helping online communities to semantically enrich folksonomies

3

… without overloading users

Page 4: Helping online communities to semantically enrich folksonomies

4

… and by collectingall user's expertiseinto the process

Page 5: Helping online communities to semantically enrich folksonomies

5

Our approach Integrate usage-analysis for a tailored solution

Supporting diverging points of view

Automatic processings +

Human expertise through user-friendly interfaces

Page 6: Helping online communities to semantically enrich folksonomies

6

Concrete scenario

Expertsproduce docs

+ tag

Archivistscentralize + tag

Public audienceread + tag

Page 7: Helping online communities to semantically enrich folksonomies

1. Modeling statements about tags

Page 8: Helping online communities to semantically enrich folksonomies

8

Supporting diverging points of view

car pollutionskos:related

Page 9: Helping online communities to semantically enrich folksonomies

9

car pollution

John Paul

Supporting diverging points of view

agrees disagrees

skos:related

Page 10: Helping online communities to semantically enrich folksonomies

Supporting diverging points of view

10

car pollutionskos:related

John Paul

hasApproved hasRejected

tagSemanticStatement (named graph)

Page 11: Helping online communities to semantically enrich folksonomies

Supporting diverging points of view

Page 12: Helping online communities to semantically enrich folksonomies

2. folksonomy enrichment Life cycle

Page 13: Helping online communities to semantically enrich folksonomies

13

Dataset

Delicious TheseNet CADIC

WhatBookmarks of users of tag

"ademe"

Keywords for Ademe's PhD

projectsArchivists

indexing lexicon

# tags 1015 6583 1439

# posts 1013 1425 4675

# (restricted tagging) 3015 10160 25515

# users 812 1425 1

Page 14: Helping online communities to semantically enrich folksonomies

ADDING TAGS

Automatic processing

User-centricstructuring

Detect conflicts

Globalstructuring

Flat folksonomy

Structured folksonomy

Folksonomy enrichment

life-cycle

Page 15: Helping online communities to semantically enrich folksonomies

15

pollution

pollutantpollution

pollutionpollutionpollutionpollution Soil pollutions

1. String-based metrics

Page 16: Helping online communities to semantically enrich folksonomies

16

Evaluation of 30 edit distances

Combining the best metrics

Needs complement !

1. String-based metrics

Page 17: Helping online communities to semantically enrich folksonomies

17

1. String-based metrics

Evaluated against Ademe expert's point of view

Page 18: Helping online communities to semantically enrich folksonomies

Related55,10%

Broader/narrower32%

CloseMatch12,82%

Result on full dataset

Page 19: Helping online communities to semantically enrich folksonomies

Result on full dataset

Node size ↔ InDegree

◉ tags (delicious + thesenet)

◉ mot-clés svic

Page 20: Helping online communities to semantically enrich folksonomies

Result on full dataset

Node size ↔ InDegree

◉ tags (delicious + thesenet)

◉ mot-clés svic

Page 21: Helping online communities to semantically enrich folksonomies

21

Fig. Markines et al. (2009)

Association via :

Users

tags

2. Tri-partite structure of folksonomies

Page 22: Helping online communities to semantically enrich folksonomies

tag1 tag2 tag3

tag1 freq (tag1) cooc (tag1, tag2) cooc (tag1, tag3)

tag2 cooc (tag2, tag1) freq (tag2) cooc (tag2, tag3)

tag3 cooc (tag3, tag1) cooc (tag3, tag2) freq (tag3)

2. Tri-partite structure of folksonomiesTag-based association :

=> gives "related" relations

Page 23: Helping online communities to semantically enrich folksonomies

Example result on CADIC dataset:

Page 24: Helping online communities to semantically enrich folksonomies

2. Tri-partite structure of folksonomies

User-based association (Mika) :

environnementagriculture

U1

U2

U3

U4

U6

=> gives "subsumption" relations

Page 25: Helping online communities to semantically enrich folksonomies

Example result on Delicious dataset:

Arrows mean "has broader"thickness ≈ weight

Page 26: Helping online communities to semantically enrich folksonomies

TheseNet dataset:

Arrows mean "has broader"thickness ≈ weight

Page 27: Helping online communities to semantically enrich folksonomies

ADDING TAGS

Automatic processing

User-centricstructuring

Detect conflicts

Globalstructuring

Flat folksonomy

Structured folksonomy

Folksonomy enrichment

life-cycle

Page 28: Helping online communities to semantically enrich folksonomies

28

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

Page 29: Helping online communities to semantically enrich folksonomies

29

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

Page 30: Helping online communities to semantically enrich folksonomies

30

Capturing user's point of view

Page 31: Helping online communities to semantically enrich folksonomies

31

Experimentation ADEME

Page 32: Helping online communities to semantically enrich folksonomies

ADDING TAGS

Automatic processing

User-centricstructuring

Detect conflicts

Globalstructuring

Flat folksonomy

Structured folksonomy

Folksonomy enrichment

life-cycle

Page 33: Helping online communities to semantically enrich folksonomies

33

Conflict detection

environment pollution

narrower

broader

Page 34: Helping online communities to semantically enrich folksonomies

34

Conflict detection

environment pollution

narrower

broader

Using rules e.g:

IF num(narrower)/num(broader) ≥ cTHEN narrower winsELSE 'more generic' wins

Page 35: Helping online communities to semantically enrich folksonomies

35

Conflict detection

environment pollution

narrower

broader

related

related

broader narrower

more generic more generic

Page 36: Helping online communities to semantically enrich folksonomies

Ademe experimentation

Total number of relations 125

Contradictory 43

Consensual 14

Only rejected (to be deleted ?) 2

Only proposed by computer (no user review) 52

Debated (approved AND rejected at least once) 14

Page 37: Helping online communities to semantically enrich folksonomies

Example result on Ademe's experimentation subset

Page 38: Helping online communities to semantically enrich folksonomies

Example result on Ademe's experimentation subset

Page 39: Helping online communities to semantically enrich folksonomies

ADDING TAGS

Automatic processing

User-centricstructuring

Detect conflicts

Globalstructuring

Flat folksonomy

Structured folksonomy

Folksonomy enrichment

life-cycle

Page 40: Helping online communities to semantically enrich folksonomies

40

environment pollutionrelated

ReferentUser

Global structuring by Referent

hasApproved

Page 41: Helping online communities to semantically enrich folksonomies

ADDING TAGS

Automatic processing

User-centricstructuring

Detect conflicts

Globalstructuring

Flat folksonomy

Structured folksonomy

Folksonomy enrichment

life-cycle

Page 42: Helping online communities to semantically enrich folksonomies

environment

pollutants

pollution

environment

pollutants

pollution

narrowernarrower

Paul

environment

pollution

narrower

John

environment

pollution

related

Referent pollutantsnarrower

Page 43: Helping online communities to semantically enrich folksonomies

43

Take away message (conclusion)

Page 44: Helping online communities to semantically enrich folksonomies

44

Help communities

structure their tags

What we do :

Page 45: Helping online communities to semantically enrich folksonomies

45

Our contributions:

Usages analysis

Automatic processing of tags

Tag structuring embedded in every-day tools

Supporting multi-points of view

Page 46: Helping online communities to semantically enrich folksonomies

46

Future work

• Interfaces :• to capture user-centric contributions • Global administrations (Referent User)• Tag searching

• Real-scale experimentation• Mapping between knowledge representation

(Gemet thesaurus – Tags – CADIC e.g.)

• Moving to other socio-structural models

Page 47: Helping online communities to semantically enrich folksonomies

47

Thank you [email protected]

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

http://isicil.inria.fr

Page 48: Helping online communities to semantically enrich folksonomies

48

Page 49: Helping online communities to semantically enrich folksonomies

49

What is a tag ?

Page 50: Helping online communities to semantically enrich folksonomies

50

Tagging model

Page 51: Helping online communities to semantically enrich folksonomies

51

Tagging model

Page 52: Helping online communities to semantically enrich folksonomies

52

Hypotheses

Tag-link is thematic

Tags are concept-candidate

Page 53: Helping online communities to semantically enrich folksonomies

53

Supporting diverging points of view

Page 54: Helping online communities to semantically enrich folksonomies

54

An eco-system of agents

Page 55: Helping online communities to semantically enrich folksonomies

55

Page 56: Helping online communities to semantically enrich folksonomies

56

Tags are nice to organize your ownresources ...

Page 57: Helping online communities to semantically enrich folksonomies

57

... but also to get involved in the organization of

shared resources