Doing Clever Things with the Semantic Web
Download as PPTX, PDF•3 likes•1,804 views
The document discusses using the Semantic Web as a knowledge base for artificial intelligence applications. It describes how the Semantic Web publishes data on the web in a standardized, linked format. This vast amount of distributed knowledge could be mined by AI in various ways, such as linking data mining to find patterns, using reasoning to analyze and understand raw data, and assessing agreement between ontologies. The Semantic Web represents a large, collaborative base of formally represented knowledge that provides many opportunities for future AI research and applications.
Doing Clever Things with the Semantic Web
- 1. Doing Clever Things With the Semantic WebMathieu d’AquinKnowledge Media Institute, the Open University
- 2. The Semantic WebUsing the Web to publish, share and exploit information/knowledgeFrom machines to machinesUsing graph-based data modeling, knowledge representation (ontologies) and reasoning
- 3. Linked Datahttp://lucero-project.info/lb/what-is-linked-data/As set of principles and technologies for a Web of DataPutting the “raw” data online in a standard representation (RDF)Make the data Web addressable (URIs)Link to other Data http://linkeddata.org
- 4. GalenNCI…MusicDCWORDNETRSSTAPFOAF…………Metadata<rdf:RDF><channel rdf:about=“http://watson.kmi.open.ac.uk/blog”><title>Elementaries - The Watson Blog</title><link>http://watson.kmi.open.ac.uk:8080/blog/</link><description>"Oh dear! Where the Semantic Web is going to go now?" -- imaginary user 23</description><language>en</language><copyright>Watson team</copyright><lastBuildDate>Thu, 01 Mar 2007 13:49:52 GMT</lastBuildDate><generator>Pebble (http://pebble.sourceforge.net)</generator><docs>http://backend.userland.com/rss</docs>…<rdf:RDF> <foaf:Imagerdf:about='http://static.flickr.com/132/400582453_e1e1f8602c.jpg'> <dc:title>Zen wisteria</dc:title> <dc:description></dc:description> <foaf:pagerdf:resource='http://www.flickr.com/photos/xcv/400582453/'/> <foaf:topicrdf:resource='http://www.flickr.com/photos/tags/vittelgarden/'/> <foaf:topicrdf:resource='http://www.flickr.com/photos/tags/wisteria/'/> <dc:creator> <foaf:Person><foaf:name>Mathieu d'Aquin</foaf:name>…<rdf:RDF> <owl:Ontology rdf:about=""> <owl:imports rdf:resource="http://usefulinc.com/ns/doap#"/> </owl:Ontology> <j.1:Organization rdf:ID="KMi"> <rdfs:comment rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >The Knoledge Media Institute of the Open University, Milton Keynes UK</rdfs:comment> </j.1:Organization> <j.1:Document rdf:ID="KMiWebSite"> …UoD
- 5. The Semantic WebKnowledge/Problem Solving MethodsSemantic Web ApplicationsDoing Clever Things With the Semantic WebIntelligent AgentSmart FeaturesClever Things…
- 6. What I Want to Talk AboutUsing the Semantic Web as A Knowledge BaseKMi Watson and Finding OntologiesDoing More with LinksExploring the Web of DataBack to the FutureAI + Linked Data = Semantic Web?
- 7. Using the Semantic WebNeed for a Gateway to the Semantic WebDynamically retrieving, exploiting and combining relevant semantic resources from the Semantic Web
- 8. KMi Watson
- 9. Architecture
- 10. Interfacehttp://watson.kmi.open.ac.ukWatson: More than a Semantic Web Search Engine, Semantic Web Journal
- 11. Watson as a ServiceProviding Web accessible APIs to a collection of online ontologies and semantic data sources
- 12. Chose an entity to searchIntegrate statements Into the edited ontologyGet entities from online ontologiesExample Application: Ontology ConstructionReusing Knowledge from the Semanrtic Web with the Watson Plugin, Demo at ISWC 2008
- 13. Concept Relation DiscoverySeaFoodMeatwine.owlAcademicStaffSemantic WebSemantic WebResearcherka2.rdfMeatSeaFoodHampizza-to-goNALTAcademicStaffResearcherHamSeaFoodISWCSWRCNALTAgrovoc
- 14. Exploring the Semantic Web as Background Knowledge for Ontology Matching, Journal of Data Semantics
- 16. From Semantic Web Research to Linked Data ApplicationsWatson as a platform to research applications and techniques on top of semantic web resourcesBut how can the Semantic Web be exploited and used in real-world application?Starting from what we know best…
- 17. Applying Linked DataThe Open University is the largest University in the UK, where all the courses are realized at a distanceCreating the first University Linked Data platform: data.open.ac.ukDemonstrate the value of the technology and push the research through real-world scenarios
- 18. data.open.ac.uk
- 19. ApplicationsMobile and Personal SemanticsSocialResource DiscoveryResearchExploration
- 20. Example application: Finding relevant resourcesZablithet al, LinkedLearning 2011
- 21. Data as Web resources, accessible everywhere
- 22. See also: Zablith et al., COLD 2011
- 24. Supporting Researchers: The Reading Experience Databasehttp://www.open.ac.uk/Arts/reading/40,000 accounts of somebody reading something at some time in some placeUsed by researchers in literature and history to explore research hypotheses
- 25. EventLocationlocatedInsubClassOfsubClassOfExperienceCityCountrydate: DatereaderInvolvedoriginCountrytextInvolvedoccupationgivesBackgroundToPersonreligiongendercreator/editorLinkedEvent OntologyDocumentCITO Citation OntologyDublin Coretitle: Stringdescription: Stringpublished: DateprovidesExcerptForFOAFDBPedia
- 27. Back to the FutureThe Semantic Web is both a vision and a realityMaking the Web more than a network of documents: the biggest, most distributed knowledge base everWhat could AI do with such a knowledge base?
- 28. Linked Data MiningFinding unexpected patterns in the use of the distributed data graph
- 29. Linked Data Mining: ExampleUsing Formal Concept Analysis + Reasoning to build a hierarchy of questions a linked dataset can answerUse statistical metrics to identify the ones that are most likely to be interestingExtracting Relevant Questions to an RDF Dataset Using Formal Concept Amalysis at KCAP 2011http://lucero-project.info/lb/2011/06/what-to-ask-linked-data/
- 32. ReasoningTo analyze and understand raw data in relation with online resourcesExample: Online personal information managementOnline Activities Ontology HTTP Ontology Parameters and Website info.Web Site InformationPersonal InformationTrust ModelLocation Information
- 33. Enriched with linked dataGoogle ServicesEntertainment WebsitesWeb AnalyticsInternet Search Enginesubject/categoryVideo sharingVideo Hostingwww.google-analytics.comCompanydeveloperWeb Search EngineSearch Enginetypesubject/categorygoogleownersubsediaryOfwww.youtube.comwww.google.comparentDBpediafreebase
- 34. Basic processing/analysisRequests by time of dayRequests by User AgentsInterestsTrust
- 36. Understanding knowledge representation and data modelingThe Semantic Web also represents a very large, collaborative base of formally represented knowledge This can also be mined, to discover things about knowledge representation and data modeling
- 37. Ontologies on the Semantic WebUnderlying description logicNumber of entitiesDomain covered
- 38. Relationships between ontologiesDOOR: Towards a Formalization of Ontology Relations at KEOD 2009
- 39. Detecting versions of ontologiesWhen published on the Web, the information about the evolution of ontologies is lostUsing URI patterns to find candidate versions of ontologieshttp://loki.cae.drexel.edu/wbs/ontology/2003/10/iso-metadata http://loki.cae.drexel.edu/wbs/ontology/2004/01/iso-metadata Applying machine learning algorithms (SVM, Naïve Bayes and Decision tree to recognize chains of versions of ontologiesAllocca at ESWC 2011Obtained 90% Precision (SVM)Collected thousands of ontology version sequences to be analysedFor example, distribution of similarity in version and non-version ontologies (right)
- 40. Agreement/Disagreement between ontologiesOntologies are knowledge artifacts, they express opinions and beliefs and contradict each othersAssessing (dis)agreement in ontologies is very useful to understand how to combine knowledge from different sources
- 41. Assessing Statements related to SeaFoodNb1: #ontologies in which the statement appears.Nb2: #ontologies containing entities matching the subject and object of the statement. a: global agreement, d: global disagreement, cs: consensus, ct: controversy
- 42. 21 different ontologies with a SeaFood conceptDisagreementAgreement Formally Measuring Agreement and Disagreement in Ontologies at K-CAP 2009
- 43. http://uciad.infoVegan subClass VegetarianSeaFoodsubClassOf MeatSeaFooddisjointWith Meat
- 44. The brighter the blue the higher the positive consensus (higher agreement)The brighter the red the lower the negative consensus (higher disagreement)Dark = controversy: no clear cut between disagreement and agreementExample: The statements attached to the class Employee are controversial: some ontologies agree, others disagree (often due to alternative representations of roles)AKT PortalUsing consensus to assess an ontology(a new NeOn toolkit pluginVisualizing Consensus with Online Ontologies to Support Quality in Ontology Development at ONTOQUAL@EKAW 2010
- 45. So my point is…The Semantic Web is a fantastic open field for AIIt is going to become omnipresent, hidden, personalExploring, Exploiting and Excavating the Semantic Web for Research in technology (creating it and studying it) Research in other areasEveryday tasks Still, after 10 years of research, represent new directions for many fields of the AI community, with their own issues, challenges and applications
Editor's Notes
- #29: Would be nice to have a bit of loted here…