20100614 ISWSA Keynote
3 likes1,792 views
The document discusses challenges facing the semantic web as it tries to keep up with the growth of the regular web, including not having enough agreed upon vocabularies, data, and links between data. It also notes problems with reasoning over large amounts of noisy and inconsistent web data from different sources. Solutions proposed include cleverly injecting semantic web technologies into content management systems to extract and link more data, as well as developing lightweight vocabularies and simplified reasoning techniques.
![Example: SIOC-2-RSSXSPARQL+SIOC enables customised RSS export:16<channel><title> {for $name from <http://www.johnbreslin.com/blog/index.php?sioc_type=site> where { [a sioc:Forum] sioc:name $name } return $name}</title> {for $seeAlso from <http://www.johnbreslin.com/blog/index.php?sioc_type=site> where { [a sioc:Forum] sioc:container_of [rdfs:seeAlso $seeAlso] } return <item> {for $title $descr $date from $seeAlso where { [a sioc:Post] dc:title $title ; sioc:content $descr; dcterms:created $date } return <title>$title</title> <description>$descr</description> <pubDate>$date</pubDate>} </item>RSS2.0“Great stuff,... I have not seen any SIOC to RSS xslt examples or vice versa” (John Breslin, creator of SIOC)](https://image.slidesharecdn.com/20100614iswsakeynote-1276394189-phpapp02/85/20100614-ISWSA-Keynote-18-320.jpg)
![RDFS+OWL inference by rules 1/2Semantics of RDFS can be partially expressed as (Datalog like) rules: rdfs1: { ?S rdf:type ?C } :- { ?S ?P ?O . ?P rdfs:domain ?C . } rdfs2: { ?O rdf:type ?C } :- { ?S ?P ?O . ?P rdfs:range ?C . } rdfs3: { ?S rdf:type ?C2 } :- {?S rdf:type ?C1 . ?C1 rdfs:subclassOf ?C2 . }cf. informative Entailment rules in [RDF-Semantics, W3C, 2004], [Muñoz et al. 2007]353535](https://image.slidesharecdn.com/20100614iswsakeynote-1276394189-phpapp02/85/20100614-ISWSA-Keynote-42-320.jpg)
![RDFS+OWL inference by rules 2/2OWL Reasoning e.g. inverseFunctionalProperty can also (partially) be expressed by Rules:owl1: { ?S1 owl:SameAs ?S2 } :- { ?S1 ?P ?O . ?S2 ?P ?O . ?P rdf:type owl:InverseFunctionalProperty }owl2: { ?Y ?P ?O } :- { ?Xowl:SameAs?Y . ?X ?P ?O }owl3: { ?S ?Y ?O } :- { ?Xowl:SameAs?Y . ?S ?X ?O }owl4: { ?S ?P ?Y } :- { ?Xowl:SameAs?Y . ?S ?P ?X }cf. pD* fragment of OWL, [ter Horst, 2005], or, more recent: OWL2 RL363636](https://image.slidesharecdn.com/20100614iswsakeynote-1276394189-phpapp02/85/20100614-ISWSA-Keynote-43-320.jpg)
20100614 ISWSA Keynote
- 1. Can Semantics catch up with the Web?Axel PolleresISWSA2010Monday, 14/06/2010Amman, Jordan
- 2. Excellent tutorial here: http://www4.wiwiss.fu- berlin.de/bizer/pub/LinkedDataTutorial/ Linked Open DataGreat!So, Can we go home and declare success?Not yet……22
- 3. 3Problem1: We’re lagging behind… From: S.Auer et al. Triplify - lightweight linked data publication from relational databases. WWW 2009.3
- 4. 4Problem2: We’re overwhelmed… After a rough estimation, it looks like the services hosted on DBTune provide access to 13.1 billion triples, therefore making a significant addition to the data web!http://blog.dbtune.org/post/2008/04/02/DBTune-is-providing-131-billion-triples…However:Full DL Reasoners choke on far less…
- 5. … they’re not made for Web Data4
- 6. 5Problem1: Too little Data… more details…HTML Web grows much faster… How to inject SW technology cleverly? … How to lift Web Data, how to reuse Semantic Web Data?Too little “agreed” vocabularies… How to build them?
- 7. Too little links/reuse … Reasoning to the rescue?5
- 8. How to inject SW technology cleverly?Example: Injecting SW Technology in Drupal6
- 9. 7Digital Enterprise Research Institutewww.deri.ieLoads of Data on the Web in CMS...7
- 10. 8Digital Enterprise Research Institutewww.deri.ieDemo site: http://drupal.deri.ie/projectblogs/So, here’s our idea of a CMS:8
- 11. Semantic Drupal:9Enables data mining techniques, text-analysis, reasoning, aggregation, trend detection over different platforms
- 12. 10Digital Enterprise Research Institutewww.deri.ieWhere is it used?Science Collaboration Framework:Stembook (Stem Cell articles and reviews)http://www.stembook.org/10
- 13. 11Digital Enterprise Research Institutewww.deri.ieISWC201011
- 14. Semantic DrupalOut-of-the-box Linked Data from any Drupal siteOut-of-the-box “site ontology”Out-of-the-box SPARQL endpointAdvanced: tie to existing vocabulariesAdvanced: import Data via SPARQLDrupal 6 modules:http://drupal.org/project/rdfcckhttp://drupal.org/project/evochttp://drupal.org/project/sparql_ephttp://drupal.org/project/rdfproxy12
- 15. 13Digital Enterprise Research Institutewww.deri.ie* http://drupal.org/project/usage/drupalGood news from Drupal 7:RDF mapping feature committed to Drupal 7 coreRDFa output by default (blogs, forums, comments, etc.)using FOAF, SIOC, DC, SKOS.Download development snapshot http://ftp.drupal.org/files/projects/drupal-7.x-dev.tar.gzCurrently more than 200.000* sites on Drupal 6waiting to make the switch to Drupal 7waiting to massively increase the amount of RDF dataon the WebHuge boost for RDF on the Web!13
- 16. 14How to lift Web Data, how to reuse Semantic Web Data?XSLT/XQueryHTMLRSS<XML/>XSPARQLSOAP/WSDLSPARQL14
- 17. 15XQuery + SPARQL = XSPARQL
- 18. Example: SIOC-2-RSSXSPARQL+SIOC enables customised RSS export:16<channel><title> {for $name from <http://www.johnbreslin.com/blog/index.php?sioc_type=site> where { [a sioc:Forum] sioc:name $name } return $name}</title> {for $seeAlso from <http://www.johnbreslin.com/blog/index.php?sioc_type=site> where { [a sioc:Forum] sioc:container_of [rdfs:seeAlso $seeAlso] } return <item> {for $title $descr $date from $seeAlso where { [a sioc:Post] dc:title $title ; sioc:content $descr; dcterms:created $date } return <title>$title</title> <description>$descr</description> <pubDate>$date</pubDate>} </item>RSS2.0“Great stuff,... I have not seen any SIOC to RSS xslt examples or vice versa” (John Breslin, creator of SIOC)
- 19. 17Problem1: Too little Data… more details…HTML Web grows much faster… How to inject SW technology cleverly? … How to lift Web Data, how to reuse Semantic Web Data?Too little “agreed” vocabularies… How to build lightweight vocabularies?
- 20. Too little links/reuse … Reasoning to the rescue?17
- 21. … How to build lightweight vocabularies? An example:Semantic Interlinking of Online Community Sites (SIOC) –Seeding a Standard18
- 22. 19 of 46
- 23. The SIOC ontologyThe main classes and properties are:20
- 24. The SIOC food chain21
- 26. Dissemination23
- 27. Another example of leveraging SW Data: SMOB
- 28. Neologism is a web-based editor for RDF Schema vocabularies and lightweight OWL ontologies.Collaborate to create and maintain vocabularies and ontologiesPublish the vocabulary on the Web according to W3C and Linked Data best practices, with views for humans (HTML, graph) and machines (RDF/XML, Turtle) Import existing vocabulariesAlso works with external namespaces(e.g., via PURL.org)Based on the popular Drupal CMSMore at http://neologism.deri.ie/25 of XYZMaking ontology building more Web-user-friendly:http://vocab.deri.ie/25
- 29. 26Problem2: We’re overwhelmed… After a rough estimation, it looks like the services hosted on DBTune provide access to 13.1 billion triples, therefore making a significant addition to the data web!http://blog.dbtune.org/post/2008/04/02/DBTune-is-providing-131-billion-triples…However:Full DL Reasoners choke on far less…
- 30. … they’re not made for Web Data26
- 31. 27Simplified “added value” proposition of Semantic Search…“explicit” dataRDF“implicit” data? Via inference usingOWL2, RDF Schema!Fig 1: RDF Web Dataset2727
- 32. Example: Finding experts/reviewers? Tim Berners-Lee, Dan Connolly, LalanaKagal, YosiScharf, Jim Hendler: N3Logic: A logical framework for the World Wide Web. Theory and Practice of Logic Programming (TPLP), Volume 8, p249-269Who are the right reviewers? Who has the right expertise?Which reviewers are in conflict? Most of the necessary data already on the Web, even as RDF! 2828
- 33. Tim BL’s FOAF file…2929
- 34. DBLP as Linked DateGives unique URIs to authors, documents, etc. on DBLP! E.g., http://dblp.l3s.de/d2r/resource/authors/Tim_Berners-Lee, http://dblp.l3s.de/d2r/resource/publications/journals/tplp/Berners-LeeCKSH08Provides RDF version of all DBLP data + query interface! 3030
- 35. Data in RDF: TriplesDBLP: <http://dblp.l3s.de/…/journals/tplp/Berners-LeeCKSH08> rdf:type swrc:Article.<http://dblp.l3s.de/…/journals/tplp/Berners-LeeCKSH08>dc:creator <http://dblp.l3s.de/d2r/…/Tim_Berners-Lee> . …<http://dblp.l3s.de/d2r/…/Tim_Berners-Lee> foaf:homepage <http://www.w3.org/People/Berners-Lee/> .…<http://dblp.l3s.de/d2r/…/Dan_Brickley> foaf:name“Dan Brickley”^^xsd:string.Tim Berners-Lee’s FOAF file:<http://www.w3.org/People/Berners-Lee/card#i>foaf:knows <http://dblp.l3s.de/d2r/…/Dan_Brickley> .<http://www.w3.org/People/Berners-Lee/card#i> rdf:type foaf:Person .<http://www.w3.org/People/Berners-Lee/card#i> foaf:homepage<http://www.w3.org/People/Berners-Lee/> .RDF Data online: Example3131
- 36. An example in SPARQL“Names of all persons who co-authored with authors of http://dblp.l3s.de/d2r/…/Berners-LeeCKSH08or known by co-authors”SELECT ?Name WHERE { <http://dblp.l3s.de/d2r/resource/publications/journals/tplp/Berners-LeeCKSH08> dc:creator ?Author. ?D dc:creator ?Author. ?D dc:creator ?CoAuthor.{ ?CoAuthor foaf:name ?Name . }UNION { ?CoAuthor foaf:knows ?Person. ?Person rdf:typefoaf:Person. ?Person foaf:name ?Name } }Doesn’t work… no foaf:knows relations in DBLP Needs Linked Data! E.g. TimBL’s FOAF file!3232
- 37. DBLP: <http://dblp.l3s.de/…/journals/tplp/Berners-LeeCKSH08> rdf:type swrc:Article.<http://dblp.l3s.de/…/journals/tplp/Berners-LeeCKSH08> dc:creator <http://dblp.l3s.de/d2r/…/Tim_Berners-Lee> . …<http://dblp.l3s.de/d2r/…/Tim_Berners-Lee> foaf:homepage <http://www.w3.org/People/Berners-Lee/> .Tim Berners-Lee’s FOAF file:<http://www.w3.org/People/Berners-Lee/card#i> foaf:knows <http://dblp.l3s.de/d2r/…/Dan_Brickley> .<http://www.w3.org/People/Berners-Lee/card#i> foaf:homepage<http://www.w3.org/People/Berners-Lee/> .33Back to the Data:Even if I have the FOAF data, I cannot answer the query:
- 38. Different identifiers used for Tim Berners-Lee
- 39. Who tells me that Dan Brickley is a foaf:Person?
- 40. Linked Data needs Reasoning!3333
- 41. The FOAF ontology… foaf:knows rdfs:domain foaf:Person Everybody who knows someone is a Person foaf:knows rdfs:range foaf:Person Everybody who is known is a Person foaf:Person rdfs:subclassOf foaf:Agent Everybody Person is an Agent. foaf:homepage rdf:type owl:inverseFunctionalProperty . A homepage uniquely identifies its owner (“key” property)… 343434
- 42. RDFS+OWL inference by rules 1/2Semantics of RDFS can be partially expressed as (Datalog like) rules: rdfs1: { ?S rdf:type ?C } :- { ?S ?P ?O . ?P rdfs:domain ?C . } rdfs2: { ?O rdf:type ?C } :- { ?S ?P ?O . ?P rdfs:range ?C . } rdfs3: { ?S rdf:type ?C2 } :- {?S rdf:type ?C1 . ?C1 rdfs:subclassOf ?C2 . }cf. informative Entailment rules in [RDF-Semantics, W3C, 2004], [Muñoz et al. 2007]353535
- 43. RDFS+OWL inference by rules 2/2OWL Reasoning e.g. inverseFunctionalProperty can also (partially) be expressed by Rules:owl1: { ?S1 owl:SameAs ?S2 } :- { ?S1 ?P ?O . ?S2 ?P ?O . ?P rdf:type owl:InverseFunctionalProperty }owl2: { ?Y ?P ?O } :- { ?Xowl:SameAs?Y . ?X ?P ?O }owl3: { ?S ?Y ?O } :- { ?Xowl:SameAs?Y . ?S ?X ?O }owl4: { ?S ?P ?Y } :- { ?Xowl:SameAs?Y . ?S ?P ?X }cf. pD* fragment of OWL, [ter Horst, 2005], or, more recent: OWL2 RL363636
- 44. RDFS+OWL inference by rules: Example:By rules of the previous slides we can infer additional information needed, e.g.TimBL’s FOAF: <…/Berners-Lee/card#i> foaf:knows <…/Dan_Brickley> . FOAF Ontology:foaf:knows rdfs:range foaf:Personby rdfs2 <…/Dan_Brickley> rdf:type foaf:Person. TimBL’s FOAF:<…/Berners-Lee/card#i> foaf:homepage<http://www.w3.org/People/Berners-Lee/> . DBLP: <…/dblp.l3s.de/d2r/…/Tim_Berners-Lee> foaf:homepage <http://www.w3.org/People/Berners-Lee/> . FOAF Ontology:foaf:homepage rdfs:type owl:InverseFunctionalProperty.by owl1 <…/Berners-Lee/card#i> owl:sameAs <…/Tim_Berners-Lee>.Who tells me that Dan Brickley is a foaf:Person? solved!
- 45. Different identifiers used for Tim Berners-Lee solved!373737
- 46. 38Web Reasoning: ChallengesScalabilityBillions or tens of billions of statements (for the moment)
- 47. Near linear scale!!!Noisy dataInconsistencies galore
- 50. 39Noisy Data: Omnipotent BeingProposition 1Web data is noisy.Proof:08445a31a78661b5c746feff39a9db6e4e2cc5cfsha1-sum of ‘mailto:’
- 51. common value for foaf:mbox_sha1sum
- 52. An inverse-functional (uniquely identifying) property!!!
- 53. Any person who shares the same value will be considered the sameQ.E.D.39
- 54. 40Noisy Data: Redefining Everything…and home in time for teaMore Proof:From http://www.eiao.net/rdf/1.0<owl:Property rdf:about="http://www.w3.org/1999/02/22-rdf-syntax-ns#type"> <rdfs:label xml:lang="en">type</rdfs:label> <rdfs:comment xml:lang="en">Type of resource</rdfs:comment> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#testRun"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#pageSurvey"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#siteSurvey"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#scenario"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#rangeLocation"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#startPointer"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#endPointer"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#header"/> <rdfs:domain rdf:resource="http://www.eiao.net/rdf/1.0#runs"/></owl:Property>Ontology hijacking!!40
- 55. 41The Web… …forecast is for muck41
- 56. 42Okay, so let’s do forward-chaining OWL 2 RL on billions of triples collected from the Web…foaf:mbox_sha1sum a owl:InverseFunctionalProperty .?xfoaf:mbox_sha1sum 08445a31a78661b5c746feff39a9db6e4e2cc5cf .OWL 2 RL rule prp-ifp: ?p a owl:InverseFunctionalProperty . ?x1 ?p ?z . ?x2 ?p ?z . ⇒ ?x1 owl:sameAs ?x2 .104?x1/?x2bindings in body108 inferred pair-wise and reflexiveowl:sameAsstatements…or in simpler terms:pow!42
- 57. 43Our Approach……pragmatic approach, making the necessary compromises… …(and some more besides)43
- 58. Apply a subset of OWL reasoning to the billion triple challenge datasetForward-chaining rule based approach, e.g.[ter Horst, 2005]Reduced output statements for the SWSE use case…Must be scalable, must be reasonable… incomplete w.r.t. OWL BY DESIGN!SCALABLE: Tailored rulesetfile-scan processingavoid joinsAUTHORITATIVE: Avoid Non-Authoritative inference(“hijacking”, “non-standard vocabulary use”)44SAOR: ScalableAuthoritative OWL Reasoner44
- 59. Scalable ReasoningScan 1: Scan all data (1.1b statements), separate T-Box statements, load T-Box statements (8.5m) into memory, perform authoritative analysis.Scan 2: Scan all data and join all statements with in-memory T-Box .Only works for inference rules with 0-1 A-Box patternsNo T-Box expansion by inference Needs “tailored” ruleset4545
- 60. Rules Applied: Tailored version of [ter Horst, 2005]46
- 61. Good “excuses” to avoid G2 rulesThe obvious:G2 rules would need joins, i.e. to trigger restart of file-scanThe interesting one:Take for instance IFP rule:Maybe not such a good idea on real Web dataMore experiments including G2, G3 rules in [Hogan, Harth, Polleres, IJSWIS 2009]4747
- 62. Authoritative ReasoningDocument D authoritative for concept C iff:C not identified by URIORDe-referenced URI of C coincides with or redirects to DFOAF spec authoritative for foaf:Person✓ MY spec not authoritative for foaf:Person✘Only allow extension in authoritative documentsmy:Person rdfs:subClassOf foaf:Person . (MY spec) ✓BUT: Reduce obscure membershipsfoaf:Person rdfs:subClassOf my:Person . (MY spec) ✘Similarly for other T-Box statements.In-memory T-Box stores authoritative values for rule executionOntology Hijacking4848
- 63. Rules AppliedThe 17 rules applied including statements considered to be T-Box, elements which must be authoritatively spoken for (including for bnode OWL abstract syntax), and output count4949
- 64. Authoritative Resoning covers rdfs: owl: vocabulary misusehttp://www.polleres.net/nasty.rdf:rdfs:subClassOf rdfs:subPropertyOf rdfs:Resource. rdfs:subClassOf rdfs:subPropertyOf rdfs:subPropertyOf. rdf:type rdfs:subPropertyOf rdfs:subClassOf. rdfs:subClassOf rdf:type owl:SymmetricProperty.Naïve rules application would infer O(n3) triples By use of authoritative reasoning SAOR/SWSE doesn’t stumble over these :rdfs :owl Hijacking5050
- 65. PerformanceGraph showing SAOR’s rate of input/output statements per minute for reasoning on 1.1b statements: reduced input rate correlates with increased output rate and vice-versa5151
- 66. ResultsSCAN 1:6.47 hrsIn-mem T-Box creation, authoritative analysis:SCAN 2:9.82 hrsScan reasoning – join A-Box with in-mem authoritative T-Box:1.925b new statements inferred in 16.29 hrsOn our agenda:More valuable insights on our experiences from Web dataG2 and G3 rules still difficult521.1b + 1.9b inferred = 3 billion triples in SWSE52
- 67. Is that enough?Well, good starting points, we believe…
- 68. … but still many open challenges…
- 69. Parallelise Reasoning [Wevaer, Hendler ISWC2009, Urbani et al. ESWC2010] … still only for RDFS or synthetic data.
- 70. Alternative approaches for Object consolidation needed, e.g. [Hogan et al. NeFoRS2010]
- 71. Query live data [Harth et al. WWW2010]
- 72. Full SPARQL querying (SPARQL 1.1)
- 73. More on Data Quality on the Web [Hogan et al. LDOW2010]53
- 74. Visit: http://pedantic-web.org/54Already several successes in finding/fixing: FOAF, dbpedia, NYtimes, even W3C specs… etc.
- 75. Linked Open DataSo, Can we go home and declare success?Not yet… But a lot of work in the right direction ongoing! ……Good: leaves us some more research to do ;-)5555
- 76. AcknowledgementsThis talk had a lot of work from different research groups in DERI:
- 77. Unit for Social Software (SIOC - John Breslin, SMOB - Alexandre Passant and their students)
- 78. Unit for Reasoning and Querying (SAOR – Aidan Hogan, XSPARQL – Nuno Lopes, Semantic Drupal – Stephane Corlosquet, Lin Clark)
- 79. Other people involved: Stefan Decker, Andreas Harth, Thomas Krennwallner, …
- 80. Thanks to all!