The Relationship between RDF and LOM in the Semantic Web

1. LOM = Words / RDF = Grammar? Nahum Korda

2. “One Standard for All: Why We Don't Want It, Why We Don't Need It” • Paraphrase from a presentation by Stephen Downes (Researcher at National Research Council, Alberta, Canada) • Presentation delivered at the Athabasca University's Edmonton office • The original argument: RDF is better because it is not restricted vocabulary (like LOM)

3. Semantic Web and Learning Objects • Could LOM be formulated (presented) in RDF? • Cold LOM be integrated with the Semantic Web? • How useful is Semantic Web for LOM?

4. Semantic Web • What is Semantic Web? • What is RDF and how is it related to XML? • What are ontologies? • Demonstration

5. World Wide Web • Markup Languages (HTML primarily) • Meta-information (tags) • Links to other files • Global hyperlinked environment

6. New Concept of Document • Not a single file • A set of related hyperlinked files • Content in various formats • Formatting information • Scripts and runtime applications

7. XSL file with formatting information Script to scroll news feeds Flash from banner exchange database XML from news feeds database XML from directory Graphic files New Concept of Document

8. New Concept of Document • A single focal point for representing all the relevant information from a wide variety of information resources • Different readability • No reading order • Information cannot be provided chronologically • Information must be simultaneously available

9. Semantic Web • Additional layers of information • Lexical definitions • Links to other relevant documents • Links to relevant people

10. Semantic Web • What is Semantic Web? • What is RDF and how is it related to XML? • What are ontologies? • Demonstration

11. Semantic Web • Generate these layers automatically • Allow software agents to manipulate this information

12. Resource Description Framework • Not just machine readable information • Also machine understandable information

13. Resource Description Framework <?xml version="1.0" ?><rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:s="http://description.org/schema/"> <rdf:Description about="http://www.w3.org/Home/Lassila" bagID="D_001"> <s:Creator>Ora Lassila</s:Creator> <s:Title>Ora's Home Page</s:Title> </rdf:Description></rdf:RDF>

14. Resource Description Framework • Subject • Predicate • Object

15. Resource Description Framework • Ora Lassila is the creator of the resource http://www.w3.org/Home/Lassila.

16.  Subject (Resource)   http://www.w3.org/Home/Lassila   Predicate (Property)   Creator  Object (literal)   "Ora Lassila" Resource Description Framework

17. Resource Description Framework

18. Semantic Web • What is Semantic Web? • What is RDF and how is it related to XML? • What are ontologies? • Demonstration

19. Ontology • System of concepts and relations between them: • Table [is piece of] furniture • Arm [is part of] body • Japan [is located in] East Asia • A method for storing and representing knowledge

20. Ontology

21. Semantic Web • Software applications can utilize meta-information for: • Annotations • Automatic summarization • Hyperlinking • Filtering

22. Notations for Logic • Conceptual Graph Interchange Form (CGIF) [Go: *x] [Person: 'John' *y] [City: 'Boston' *z] [Bus: *w] (Agnt ?x ?y) (Dest ?x ?z) (Inst ?x ?w) • Knowledge Interchange Format(KIF) (exists ((?x Go) (?y Person) (?z City) (?w Bus)) (and (Name ?y 'John) (Name ?z 'Boston) (Agnt ?x ?y) (Dest ?x ?z) (Inst ?x ?w)))

23. RDF and Ontologies • DARPA Agent Markup Language (DAML) • Ontology Interchange Language (OIL) • An extension to XML and RDF

24. DAML+OIL <rdfs:Class rdf:ID="BigFoot"> <rdfs:comment> BigFoots (BigFeet?) are exactly those persons whose shosize is over12. </rdfs:comment> <daml:intersectionOf rdf:parseType="daml:collection"> <rdfs:Class rdf:about="#Person"/> <daml:Restriction> <daml:onProperty rdf:resource="#shoesize"/> <daml:hasClass rdf:resource="http://www.w3.org/TR/2001/NOTE- daml+oil-walkthru-20011218/daml+oil-ex-dt#over12"/> </daml:Restriction> </daml:intersectionOf> </rdfs:Class>

25. Problems with Ontologies • Development and maintenance of ontologies is extremely complex and resource consuming • A useful ontology is domain specific and many various ontologies need to be developed in order to cover a variety of possible interests • Ontologies suffer from problems related to the intrinsic complexity

26. Intrinsic Complexity • Concepts in ontologies are often highly ambiguous • Ontologies must often handle multi-lingual document corpora • Criteria for defining interrelationships are typically highly inconsistent

27. Ambiguity • Weather in October is unstable. • Red October was one of the most important historical events in the 20th Century. • This was the first time that American military personnel came aboard RedOctober. Month in Calendar Historical Event Imaginary Naval Vessel

28. Multilingualism

29. Mathematical Ontology is a subtype of is a subclass of Circle Circle Ellipse Ellipse Object-Oriented Programming Language Inconsistency

30. Automated Idiomatic Representation • Words are not mathematical variables • Meaning is generated on the pragmatic (contextual) level – not semantic! • The meaning of a word is its use in the language. (Ludwig Wittgenstein, Philosophical Investigations, 43)

31. Automated Idiomatic Representation

32. Semantic Web • What is Semantic Web? • What is RDF and how is it related to XML? • What are ontologies? • Demonstration

33. Loads of Buzz – Little to Show • Dublin Core Metadata - DC 15 core metadata categories can be expressed in RDF in web pages • MusicBrains metadata initiative - storing and exchanging metadata related to digital audio and video tracks • ELSST thesaurus - EU thesaurus project drafted RDF format for ISO multilingual thesauri standard and the ELSST multi-lingual thesaurus of social science terms encoded in RDF

34. Loads of Buzz – Little to Show • Annotea and Mozilla support RDF annotation servers that allow users to annotate web pages • Open Directory Project (DMOZ) - a hand crafted web directory that uses RDF to store the details of web sites

35. Semantic Web and Learning Objects • Could LOM be formulated (presented) in RDF? • Cold LOM be integrated with the Semantic Web? • How useful is Semantic Web for LOM?

36. How useful is Semantic Web for LOM? • Discovery • Search more efficiently • Apply automation to information retrieval • Interoperability • Automate reusability • Bridge multilingualism

37. Cold LOM be integrated with the SW? • Restricted vocabulary necessary in order to make it interoperable • Restricted vocabulary imposes severe limitations to expressibility

38. Restricted Vocabulary - Necessity • [Go: *x] [Person: 'John' *y] [City: 'Boston' *z] [Bus: *w] (Agnt ?x ?y) (Dest ?x ?z) (Inst ?x ?w) • [Go: *x] [Person: 'John' *y] [City: 'Boston' *z] [Bus: *w] (Actr ?x ?y) (Dest ?x ?z) (Inst ?x ?w) • Agnt = Actr ? • Agnt ⊂ Actr ? • Agnt ⊃ Actr ? • Agnt ⋂ Actr ?

39. Restricted Vocabulary - Limitation • DE PROFUNDIS • There is a stubble field on which a black rain falls. There is a tree which, brown, stands lonely here. There is a hissing wind which haunts deserted huts - how sad this evening. Past the village pond the gentle orphan still gathers scanty ears of corn. Golden and round her eyes are gazing in the dusk and her lap awaits the heavenly bridegroom. Returning home shepherds found the sweet body decayed in the bramble bush. A shade I am remote from somber hamlets. The silence of God I drank from the woodland well. On my forehead cold metal forms. Spiders look for my heart. There is a light that fails in my mouth. At night I found myself upon a heath, thick with garbage and the dust of stars. In the hazel copse crystal angels have sounded once more. • Georg Trakl (1887-1914)

40. Restricted Vocabulary - Limitation • DE PROFUNDIS • There is a stubble field on which a black rain falls. There is a tree which, brown, stands lonely here. There is a hissing wind which haunts deserted huts - how sad this evening. Past the village pond the gentle orphan still gathers scanty ears of corn. Golden and round her eyes are gazing in the dusk and her lap awaits the heavenly bridegroom. Returning home shepherds found the sweet body decayed in the bramble bush. A shade I am remote from somber hamlets. The silence of God I drank from the woodland well. On my forehead cold metal forms. Spiders look for my heart. There is a light that fails in my mouth. At night I found myself upon a heath, thick with garbage and the dust of stars. In the hazel copse crystal angels have sounded once more. • Georg Trakl (1887-1914)

41. Language • Words are not mathematical variables • Meaning is generated on the pragmatic (contextual) level – not semantic? • The meaning of a word is its use in the language. (Ludwig Wittgenstein, Philosophical Investigations, 43)

42. Could LOM be formulated in RDF? • Sure – everything can be (somehow) formulated in RDF. • Could LOM be effectively/usefully formulated in RDF? • Only if the “restricted vocabulary conflict” is successfully solved • This is however true of every Semantic Web application

43. Thank You Nahum Korda