URI Disambiguation in the Context of Linked Data
- 1. http://dbpedia.org/resource/Tim_Berners-Leehttp://dbpedia.org/resource/Spainhttp://acm.rkbexplorer.com/id/resource-P112732URI Disambiguation in the Context of Linked Datahttp://sws.geonames.org/2510769http://acm.rkbexplorer.com/id/person-282197http://id.ecs.soton.ac.uk/person/7113http://www.w3.org/People/Berners-Lee/card#ihttp://id.ecs.soton.ac.uk/person/21http://www4.wiwiss.fu-berlin.de/dblp/resource/person/100007http://citeseer.rkbexplorer.com/id/resource-CSP109020http://southampton.rkbexplorer.com/id/person-00021http://www4.wiwiss.fu-berlin.de/factbook/resource/Spain
- 2. URI Disambiguation in the Context of Linked DataPresentation OutlineLinked Data RepositoriesCoreference on the Semantic WebAuthor DisambiguationDBLP Linked DataDBLP Author DisambiguationDisambiguation ResultsDBpediaPossible SolutionsSummaryLDOW2008 - Beijing, China2
- 3. URI Disambiguation in the Context of Linked DataRKBexplorer.comContains URIs for more than 10 million entitiesOver 25 Linked Data sites, including:Data relating to people, projects, papers and institutionsA single entity has a number of URIs (even within the same repository)Entities are linked using CRSesLDOW2008 - Beijing, China3DBLP
- 4. URI Disambiguation in the Context of Linked DataLinked Data RepositoriesExisting databases on the Web are being exposed as Linked Data (D2R, Virtuoso)Databases contain inconsistencies and require constant curationDatasets such as Wikipedia are being continually checked and updated, especially in the case of disambiguation (WikiProject_Disambiguation)Linked Data repositories should also provide consistent dataLDOW2008 - Beijing, China4
- 5. URI Disambiguation in the Context of Linked DataDisambiguation on the Semantic WebCoreference on the Semantic Web is defined as being the situation where two or more URIs are used for a single non-information resourceURI usage can change with contextNon-Information resource equality is hard to define preciselyExamples‘Hugh Glaser’ at Southampton vs. ‘Hugh Glaser’ at Imperial‘Harry Potter and the Order of the Phoenix’ in Hardback vs. Softback ISBN: 978-0747561071 978-07475510035LDOW2008 - Beijing, China
- 6. URI Disambiguation in the Context of Linked DataURI MultiplicityURIs for ‘Spain’:http://dbpedia.org/resource/Spainhttp://ww4.wiwiss.fu-berlin.de/factbook/resource/Spainhttp://sws.geonames.org/2510769http://www4.wiwiss.fu-berlin.de/eurostat/resource/countries/Espa%C3%BlaURIs for ‘Hugh Glaser’:http://acm.rkbexplorer.com/id/resource-P112732 http://citeseer.rkbexplorer.com/id/resource-CSP109020 http://citeseer.rkbexplorer.com/id/resource-CSP109013 http://citeseer.rkbexplorer.com/id/resource-CSP109011 http://citeseer.rkbexplorer.com/id/resource-CSP109002 http://dblp.rkbexplorer.com/id/resource-27de9959 http://europa.eu/People/#person-0ff816fa http://resist.ecs.soton.ac.uk/wiki/User:hugh_glaser http://id.ecs.soton.ac.uk/people/21 6LDOW2008 - Beijing, China
- 7. URI Disambiguation in the Context of Linked DataAuthor DisambiguationA known problem in the Information Science fieldHow to determine:Hugh Glaser/H. Glaser/Glaser, H. are the same person?How to determine:Tom Anderson – Newcastle UniversityTom Anderson – University of Washington are different people?7LDOW2008 - Beijing, China
- 8. URI Disambiguation in the Context of Linked DataExisting ApproachesString Metrics- Name Equivalence identification- Record Linkage- Citation MatchingWeb Assisted- Look up publications on author’s home page- Use search engine results on publication titleMachine Learning- k-way spectral clustering- Use author name, co-author frequency and publication venue8LDOW2008 - Beijing, China
- 9. URI Disambiguation in the Context of Linked DataDBLP Linked DataConverted from an XML dump of DBLP database950 000 Publications540 000 Authors28 million triplesUpdated WeeklyLinked to other datasets including RDF Book Mashup and RKBExplorer.com9LDOW2008 - Beijing, China
- 10. URI Disambiguation in the Context of Linked DataDBLP Author Disambiguation49 names - 10 most common English surnames with 5 common first namesAuthors disambiguated by looking at homepage, web publication, search engine results and institutionWhen in doubt, authors assumed to be the same if:- The co-authors of any publication are the same- The publication venue was the same- The area of research was the same10LDOW2008 - Beijing, China
- 11. 8LDOW2008 – Beijing, ChinaURI Disambiguation in the Context of Linked DataIt’s all about IdentityTom Anderson – http://www4.wiwiss.fu-berlin.de/dblp/resource/person/109074Is dc:creator of <http://www4.wiwiss.fu berlin.de/dblp/resource/record/conf/dac/MorettiHNCKABDF01> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/ftcs/SaeedLA91>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/ftrtft/LemosSA92>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/hybrid/AndersonLFS92>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/iccbss/AndersonFRR03>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/iciap/TruccoARI05>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/icnp/ElySWSA01> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/ifip/AndersonRR04>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/sc/BorchersASW95>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/seaai/AndersonH98> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/srds/Anderson86>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/words/AndersonFRR05>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/bell/LiuBFSRA04> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/cj/LemosSA92>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/dt/Anderson01>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/dt/Anderson03> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/dt/ZorianASTI96> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/software/LemosSA95> is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/ton/SavageWKA01>is dc:creator of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/journals/tse/AndersonBHM85> is dblp:editor of <http://www4.wiwiss.fu-berlin.de/dblp/resource/record/conf/sigcomm/2006>Vice President O-in Design Automation inc. USAProfessor, University of NewcastleProfessor, Heriot Watt UniversityUniversity of WashingtonUniversity of California, BerkelyTom Andersen - University of DenmarkLucent Technologies, Illinois
- 12. URI Disambiguation in the Context of Linked DataDBLP Author Disambiguation Results92% of authors with common names had publications incorrectly mergedWorst case - 15 different authors with 1 URIMany authors who are the same have publications under different names (Cliff Jones, C.B. Jones)Inconsistency in data means inconsistency with linked dataIt is incorrect to use owl:sameAs to link different authors who have the same URI12LDOW2008 - Beijing, China
- 13. URI Disambiguation in the Context of Linked DataDBpediaDBpedia 3.0 improves disambiguation management by including the ‘disambiguates’ propertyowl:sameAs linkage still inconsistent: <http://dbpedia.org/resource/Welsh > owl:sameAs <http://sw.cyc.com/2006/07/27/cyc/EthnicGroupOfWelsh> . <http://sw.cyc.com/2006/07/27/cyc/Welsh-TheWord> . <http://sw.cyc.com/2006/07/27/cyc/WelshLanguage> . <http://sw.cyc.com/2006/07/27/cyc/Welshing-Cheating> .<http://dbpedia.org/resource/H.P._Lovecraft> owl:sameAs <http://sw.cyc.com/2006/07/27/cyc/HPLovecraft-Author> . <http://zitgist.com/music/artist/8047a401-5ca7-48dd-9d7c-2d2b822e51e6> .13LDOW2008 - Beijing, China
- 14. URI Disambiguation in the Context of Linked DataPossible SolutionsCRS: Consistent Reference Service- Groups similar URIs into ‘bundles’- Bundles can be made according to context- Each KB can have one or more CRSesOKKAM- Coming up soon!14LDOW2008 - Beijing, China
- 15. URI Disambiguation in the Context of Linked DataSummaryLinked Data providers need to think about data consistency in the same way as database providersFailure to manage coreference within datasets leads to incorrect linkage with other datasetsThe network effect of the Web of Data means coreference needs to be even more carefully managed than in the Web of DocumentsSystems are being developed to help manage coreference, the community needs to decide how to handle the problem15LDOW2008 - Beijing, China
- 16. URI Disambiguation in the Context of Linked DataQuestions?Further questions:a.o.jaffrihg @ecs.soton.ac.ukicm16LDOW2008 - Beijing, China