semantic and social (intra)webs
Upcoming SlideShare
Loading in...5
×
 

Like this? Share it with your network

Share

semantic and social (intra)webs

on

  • 1,800 views

ICEIS 2013 Keynote talk on semantic and social (intra)webs

ICEIS 2013 Keynote talk on semantic and social (intra)webs

Statistics

Views

Total Views
1,800
Views on SlideShare
1,800
Embed Views
0

Actions

Likes
1
Downloads
44
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

CC Attribution License

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

semantic and social (intra)webs Presentation Transcript

  • 1. semantic and social (intra)webs fabien gandon, @fabien_gandon, http://fabien.info leader of the Wimmics research team (INRIA, CNRS, Univ. Nice) W3C Advisory Committee Representative for INRIA
  • 2. eb- nstrumented an- achine nteractions, ommunities, and emantics a joint research team between INRIA Sophia Antipolis – Méditerranée and I3S (CNRS and University Nice Sophia Antipolis).
  • 3. membersHead (and INRIA contact): Fabien Gandon Vice Head (and I3S contact): Catherine Faron-Zucker Researchers: 1. Michel Buffa, MdC (UNS) 2. Olivier Corby, CR1 (INRIA) 3. Alain Giboin, CR1 (INRIA) 4. Nhan Le Thanh, Pr. (UNS) 5. Isabelle Mirbel, MdC, HDR (UNS) 6. Peter Sander, Pr. (UNS) 7. Andrea G. B. Tettamanzi, Pr. (UNS) 8. Serena Villata, RP (INRIA) Post-doc: 1. Zeina Azmeh (I3S) 2. Elena Cabrio (CORDIS) Research engineers: 1. Julien Cojan (INRIA, Ministry of Culture) 2. Christophe Desclaux (Boost your code) 3. Amosse Edouard (I3S) PhD students: 1. Pavel Arapov, 3rd year (EDSTIC-I3S) 2. Franck Berthelon, 4th year (UNS-EDSTIC) 3. Ahlem Bouchahda, 3rd year (UNS-SupCom Tunis) 4. Khalil Riad Bouzidi, 3rd year (UNS-CSTB) 5. Luca Costabello, 3rd year (INRIA-CORDI) 6. Papa Fary Diallo, 2nd year (AUF-UGB-INRIA) 7. Corentin Follenfant, 3rd year (SAP) 8. Zide Meng, 1st year (INRIA ANR OCTOPUS) 9. Maxime Lefrançois, 3rd year (EDSTIC-INRIA) 10. Nguyen Thi Hoa Hue, 2nd year (Vietnam-CROUS) 11. Nicolas Marie, 3rd year (Bell-ALU, INRIA) 12. Rakebul Hasan, 2nd year (INRIA ANR-Kolflow) 13. Oumy Seye, 2nd year, (INRIA Rose Dieng allocation) Assistants: • Christine Foggia (INRIA) • Magali Richir (I3S)
  • 4. research problem socio-semantic networks: combining formal semantics and social semantics on the web
  • 5. web landscape and graphs (meta)data of the relations and the resources of the web …sites …social …of data …of services + + + +… web… = + …semantics + + + +…= + typed graphs web (graphs) networks (graphs) linked data (graphs) workflows (graphs) schemas (graphs)
  • 6. challengesanalyzing, modeling, formalizing and implementing graph-based social semantic web applications for communities  multidisciplinary approach for analyzing and modeling the many aspects of intertwined information systems communities of users and their interactions  formalizing and reasoning on these models new analysis tools and indicators new functionalities and better management
  • 7. diffusion Master IFI: from KIS to Web – gradual changes to the courses – then replace the master by a new one Standardization participation – Working groups: RDF 1.1, SPARQL 1.1 – INRIA Advisory Committee Representative Open-source and CeCILL-C free software
  • 8. CORESE/ KGRAM
  • 9. projectsisicil.inria.fr (ANR)  enterprise social networking  business intelligence, watching, monitoring  communities of interest, of practice, of experts datalift.org (ANR)  from raw public data to interlinked data and schemas  a platform and documentation to assist the process  validation on real datasets kolflow.univ-nantes.fr (ANR)  reduce the overhead of communities building knowledge  federated semantic: distributed blackboard for man-machine coop. dbpedia.fr (Ministry of Culture)  extract and publish data and facts from French version of wikipedia ocktopus (ANR)  dynamic typed network of user-generated web sites (e.g. forums) DATALIFT ...
  • 10. ISICILsemantic and social intraweb for corporate intelligence and watch ANR project CONTINT 2009-2011
  • 11. <background_knowledge>
  • 12. based on Clay Shirky, TED 2009 1↔1 1→n 1→n 1→n
  • 13. read-write web internet n↔n classical web 1↔n wiki, (µ)blog, forum, etc. n↔n Ward Cunningham, 94
  • 14. social web networks
  • 15. social networks networking is not that new e.g. commerce social network analysis beginning of the 20th century
  • 16. collaboratively create and manage tags to annotate and categorize content SOCIAL TAGGING
  • 17. a crowd of users creating massive categorizations U T R
  • 18. semantic web mentioned by Tim BL in 1994 at WWW [Tim Berners-Lee 1994, http://www.w3.org/Talks/WWW94Tim/]
  • 19. W3C®SEMANTIC WEB STACK
  • 20. W3C® A WEB OF LINKED DATA
  • 21. RDFstands for Resource: pages, images, videos, ... everything that can have a URI Description: attributes, features, and relations of the resources Framework: model, languages and syntaxes for these descriptions
  • 22. RDFis a triple model i.e. every piece of knowledge is broken down into ( subject , predicate , object )
  • 23. doc.html has for author Fabien and has for theme Music
  • 24. doc.html has for author Fabien doc.html has for theme Music
  • 25. ( doc.html , author , Fabien ) ( doc.html , theme , Music ) ( subject , predicate , object )
  • 26. RDFtriples can be seen as arcs of a graph (vertex,edge,vertex)
  • 27. Fabien author doc.html theme Music
  • 28. http://ns.inria.fr/fabien.gandon#me http://inria.fr/schema#author http://inria.fr/rr/doc.html http://inria.fr/schema#theme "Music"
  • 29. principles  use RDF as data format  use URIs as names for things  use HTTP URIs so that people can look up those names  when someone looks up a URI, provide useful information (RDF, HTML, etc.) using content negotiation  include links to other URIs so that related things can be discovered
  • 30. May 2007 April 2008 September 2008 March 2009 September 2010 Linking Open Data Linking Open Data cloud diagram, by Richard Cyganiak and Anja Jentzsch. http://lod-cloud.net/ September 2011 0 100 200 300 400 10/10/2006 28/04/2007 14/11/2007 01/06/2008 18/12/2008 06/07/2009 22/01/2010 10/08/2010 26/02/2011 14/09/2011 01/04/2012
  • 31. query with SPARQL SPARQL Protocol and RDF Query Language
  • 32. e.g. DBpedia
  • 33. -> Ontology ontologies
  • 34. W3C® PUBLISHED SEMANTICS OF SCHEMAS
  • 35. RDFS to declare classes of resources, properties, and organize their hierarchy Document Report creator author Document Person
  • 36. OWLin one… enumeration intersection union complement  disjunction restriction! cardinality 1..1 algebraic properties equivalence [>18] disjoint union value restrict. disjoint properties qualified cardinality 1..1 ! individual prop. neg chained prop.   keys …
  • 37. SKOS knowledge thesauri, classifications, subjects, taxonomies, folksonomies, ... controlled vocabulary 40
  • 38. natural language expressions to refer to concepts 41 inria:CorporateSemanticWeb skos:prefLabel "corporate semantic web"@en; skos:prefLabel "web sémantique d'entreprise"@fr; skos:altLabel "corporate SW"@en; skos:altLabel "CSW"@en; skos:hiddenLabel "web semantique d'entreprise"@fr. labels
  • 39. between conceptsinria:CorporateSemanticWeb skos:broader w3c:SemanticWeb; skos:narrower inria:CorporateSemanticWiki; skos:related inria:KnowledgeManagement. relations
  • 40. </background_knowledge>
  • 41. Social web Semantic web Linked data Social semantic web ISICIL: semantic social intraweb
  • 42. isicil.inria.fr • enterprise social networking • business intelligence, watching, monitoring • communities of interest, of practice, of experts
  • 43. proposed overview… integrating requirement analysis methods examples of challenges and derived functionalities overview of this open-source platform http://isicil.inria.fr
  • 44. MERGING METHODOLOGIES extracts of the requirement analysis and specifications
  • 45.  analyze and model key business processes  Analyze interactions between members of the group ADEME « roadmap for urban mobility »  campaigns of questionnaires at Orange Labs  trend analysis of intelligence market and watch  comparison of the APIs, widgets and other applications usage analysis an specification
  • 46. ex. study of transformations: existing  target
  • 47. convergence matrix detections of needs or redundancies in key scenarios Etapes des scénarios Fonctionnalités identifiées Fonctions SI Présenter problématique au SVIC Mailing, Q&A Envoyer Demander ce qui est incontournable et ce que font les autres ingénieurs Consultation d’experts Extraire, filtrer Prendre en compte demande Workflow, Outils de collaboration communiquer Préparer requêtes Moteur de recherche, équation de recherche rechercher Recueillir résultats Abonnement, push… Extraire, annoter Vérifier pertinence des résultats Analyses, outils de filtrage filtrer Informer l'ingénieur Messagerie électronique, chat, vidéo-conférence envoyer S'approprier les résultats et les requêtes Equation de recherche, profil, tags Annoter, organiser Devenir le destinataire des alertes Diffusion par profil diffuser
  • 48. business intelligence market analysis
  • 49. proposing functionalities & prototypes Prioritization of functionalitiesFrequent functionalities and dependencies
  • 50. EXAMPLES OF FONCTIONNALITIES examples of challenges and implementations (web 3.0 & enterprise 3.0)
  • 51. assited structuring of folksonomies flat folksonomiesweb 2.0 [Limpens et al.] pollution soil pollution has narrower pollutant energy related related thesaurus ? SKOS
  • 52. global giant graph link users, actions, knowledge, resources, groups, etc. #Freddy #bk81 hasBookmark hasTag #tag27 industry hasLabel #Fabien #bk34 #tag92 industries hasBookmark hasLabel hasTag
  • 53. pollution pollutantpollution pollutionpollutionpollutionpollution Soil pollutions edition distances
  • 54. 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 Close match Aire sous f cm_A Tag1 broader than Tag2 Aire sous f t1bt2_A Related Aire sous f rel_A evaluating distancesc.f. [Limpens et al.]
  • 55. determine thesaurus relations Comparison of the mean value of the JaroWinkler metric for each type of semantic relation Mean value of the difference s(t1,t2) - s(t2,t1) with s being the Monge-Elkan QGram metric for each set of tag pairs.
  • 56. Node size ↔ InDegree ◉ tags (delicious + thesenet) ◉ svic keywords examples of results
  • 57. 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) contextual distance: co-occurrence vector cosine distance to detect related tags   21 21 21,cos tagtag tagtag tagtag    [Cattuto et al. 2008]
  • 58. example of resultsCADIC, ADEME
  • 59. environment interest comunity inclusion detecting narrower tags [Mika et al.] agriculture      12 ,,21 tagrhasnarrowetaguseruser tagtag 
  • 60. Arrows mean "has broader" thickness ≈ weight examples of resultsdel.icio.us
  • 61. combining metrics edition distances Monge-Elkan Soundex, JaroWinkler, asymmetry Monge-Elkan Qgram contextual metric cosinus vector co-occurring tags social metrics inclusion of communities of interest football sport + +
  • 62. 83 027 relations / 9 037 tags  68 633 related  11 254 hyponyms  3 193 spelling variants
  • 63. structuring as a side effect
  • 64. handling conflicts arbitration rules IF num(narrower)/num(broader) ≥ c THEN narrower/broader ELSE related purely automatic conflicting arbitrated conflict debated consensual
  • 65. folksonomy enrichment lifecycle ADDING TAGS Automatic processing User-centric structuring Detect conflicts Global structuring Flat folksonomy Structured folksonomy [Limpens et al.]
  • 66. Graphs, graphs, graphs Fabien owner author Researcher type doc.html author Semantic web is not antisocial Adult Researcher sub property sub class semantic web title Fabien Marco Guillaume Nicolas Michel Rémi social network analysis  ),(;)( pxrelxpdin  4)(  Guillaumedin owner Adult type
  • 67. parentsibling motherfatherbrothersister colleague knows Gérard Fabien Mylène Michel Yvonne <family> (guillaume)=5d(guillaume)=3guillaume c.f. [Erétéo et al.]
  • 68. eg. typed proximity centrality select distinct ?y ?to pathLength($path) as ?length (1/sum(?length)) as ?centrality where{ ?y s (foaf:knows*/rel:worksWith)::$path ?to }group by ?y      1           GEx worksWithknows c worksWithknows xkglengthkC ,/*/*
  • 69. e.g.
  • 70. ipernity.com dataset in RDF 61 937 actors & 494 510 relationships –18 771 family links between 8 047 actors –136 311 friend links implicating 17 441 actors –339 428 favorite links for 61 425 actors etc. c.f. [Erétéo et al.]
  • 71. some interpretations validated with managers of ipernity.com friendOf, favorite, message, comment small diameter, high density family as expected: large diameter, low density favorite: highly centralized around Ipernity animator. friendOf, family, message, comment: power law of
  • 72. some interpretations existence of a largest component in all sub networks "the effectiveness of the social network at doing its job" [Newman 2003] 0 10000 20000 30000 40000 50000 60000 70000 number actors size largest component knows favorite friend family message comment
  • 73. e.g. of results: different key actors for different kinds of links c.f. [Erétéo et al.]
  • 74. PERFORMANCES & LIMITS Knows 0.71 s 494 510 Favorite 0.64 s 339 428 Friend 0.31 s 136 311 Family 0.03 s 18 771 Message 1.98 s 795 949 Comment 9.67 s 2 874 170 Knows 20.59 s 989 020 Favorite 18.73 s 678 856 Friend 1.31 s 272 622 Family 0.42 s 37 542 Message 16.03 s 1 591 898 Comment 28.98 s 5 748 340 Shortest paths used to calculate Knows Path length <= 2: 14m 50.69s Path length <= 2: 2h 56m 34.13s Path length <= 2: 7h 19m 15.18s 100 000 1 000 000 2 000 000 Favorite Path length <= 2: 5h 33m 18.43s 2 000 000 Friend Path length <= 2: 1m 12.18 s Path length <= 2: 2m 7.98 s 1 000 000 2 000 000 Family Path length <= 2 : 27.23 s Path length <= 2 : 2m 9.73 s Path length <= 3 : 1m 10.71 s Path length <= 4 : 1m 9.06 s 1 000 000 3 681 626 1 000 000 1 000 000 )(GComp rel )(, yD rel  1 time projections
  • 75. SEMSNA SCHEMA annotating the networks with analysis results high centrality
  • 76. SemSNA an ontology of SNA http://ns.inria.fr/semsna/
  • 77. 4 Philippe colleague 2 colleague supervisor Degree Guillaume Gérard Fabien Mylène Michel Yvonne IvanPeter example of SemSNA
  • 78. ADD { ?y semsna:hasInDegree _:b0 _:b0 semsna:forProperty param[type] _:b0 rdf:value ?indegree _:b0 semsna:hasLength param[length] } SELECT ?y count(?x) as ?indegree { ?x $path ?y filter(match($path, star(param[type]))) filter(pathLength($path)<= param[length]) } group by ?y parameterized in-degree )(do lengthtype, yin  [PhD Guillaume Erétéo]
  • 79. conceptual and software framework for a semantic analysis of social networks using semantic web frameworks
  • 80. groups&reasons
  • 81. detecting AND labeling communities ? ?
  • 82. propagating tags to discover communities of interest
  • 83. tags to detect and label communities extension of algorithm RAK/LP : from random labels to structured tags salt, water pepper, wine mustard rugby, foot foot, movie hockey sport sport sport condiment condiment condiment [Eretéo et al., 2011]
  • 84. applied to Ademe Ph.D. network  1 853 agents 1 597 academic supervisors 256 ADEME engineers.  13 982 relationships 10 246 rel:worksWith 3 736 rel:colleagueOf  6 583 tags  3 570 skos:narrower relations between 2 785 tags
  • 85. -1,2 -1,1 -1 -0,9 -0,8 -0,7 -0,6 -0,5 -0,4 -0,3 -0,2 -0,1 0 0,1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 R AK TagP S emTagP controlled S emTagP (env) controlled S emTagP (env, energ) controlled S emTagP (env, energ, model) MODULARITY COMPARISONS X axis: propagation iterations, Y axis: modularity
  • 86. results 1. pollution 2. sustainable development 3. energy 4. chemistry 5. air pollution 6. metals 7. biomass 8. wastes 92
  • 87. controled abstraction and merge lead
  • 88. PLATFORM AND PROTOTYPES overview of the architecture
  • 89. a web 3.0 solution open integration and standard in the front-end
  • 90. towardsrichwebmarks
  • 91. navigating in the expert network
  • 92. activity flow and notification
  • 93. Semantic Web plugin for Gephi
  • 94. web-scraping: archiving and integrating
  • 95. Fresnel lenses to adapt results
  • 96. create dynamic reports in the wiki
  • 97. integrating with new web platforms
  • 98. going mobile
  • 99. to know more deliverables and publications http://isicil.inria.fr open source code on INRIA forge https://gforge.inria.fr/projects/isicil/ models http://ns.inria.fr/
  • 100. Xxxxx xxx Xxxxx xxxx xxxx xxxxx xxxx Xxxxx x xxxxx gave birth to … « Unveil the semantic imprint from within your Community » • cooperative society of social semantic web specialists • industrialization and maintenance of research results • developing webwarks as linked semantic traces Xxxxxx xxxxxx xxxxxxXxxxxx xxxxxx xxxxxx Xxxxxx xxxxxx xxxxxx contribute downloaded interested in exchange Xxxxxx xxxxxx xxxxxxcontributes exhange interested in contributes Xxxxxx xxxxxx xxxxxx interested in Xxxxxx xxxxxx xxxxxx link & enrich analyse & assist
  • 101. doggy-bag of the talk
  • 102. web 1, 2
  • 103. price convert? person homepage? more info? web 1, 2, 3
  • 104. wikipedia editions by users… 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 without bots…
  • 105. semantic spreading activation some other contributions mappings linguistic patterns [Person] is starring in [Movie] [Person] born in [Place] (…) exploratory search natural language Q&A publication & links projection & inferences ASK { ?res creator ?prov . ?prov hasColleague ?user . ?prov interestedBy ?topic . ?user interestedBy ?topic }
  • 106. informal formal usage representation one web… data person document program metadata
  • 107. tomorrow, he who controls the metadata, controls the web. @fabien_gandon http://fabien.info