The Semantic Web, Ontologies and Beyond - PowerPoint PPT Presentation

1 / 17
About This Presentation
Title:

The Semantic Web, Ontologies and Beyond

Description:

These are not sufficient for more context sensitive and semantic ... Pellet, Racer etc. Ontology Integration Techniques: ontology mapping, alignment, merging ... – PowerPoint PPT presentation

Number of Views:32
Avg rating:3.0/5.0
Slides: 18
Provided by: SAN74
Category:

less

Transcript and Presenter's Notes

Title: The Semantic Web, Ontologies and Beyond


1
The Semantic Web, Ontologies and Beyond
  • Albert Bokma
  • Research Fellow
  • University of Sunderland
  • Albert.bokma_at_sunderland.ac.uk

2
Current Developments
  • 1.5 billion users
  • more than 10 billion pages
  • Exponential growth pattern

WWW URL, HTML, HTTP
3
Evolution of the Web
Semantic Web Services OWL-S, WSMO
Semantic Web RDF, RDF(S), OWL
Web Services UDDI, WSDL, SOAP
WWW URL, XHTML, HTTP
4
The Semantic Web
  • XHTML - The standard web uses display oriented
    markup
  • XML - Content oriented markup is available as a
    format for data exchange
  • These are not sufficient for more context
    sensitive and semantic processing
  • OWL, RDF and RDF(S) the semantic markup based
    on ontologies
  • Used as separate or additional markup
  • Backward compatibility
  • Machine processable semantics

5
So what is an ontology?
  • According to the classical definition, an
    ontology is
  • A formal, explicit specification of a shared
    conceptualisation
  • This provides us with a modelling framework with
    the following attributes
  • Formal means that it is unambiguous
  • Explicit means that it is machine processable
  • Shared means that we can effectively communicate

6
Available Technologies
  • A host of Semantic Web tools and technologies are
    available
  • Ontology Languages
  • RDF, RDF(S) and OWL etc.
  • Ontology Development Tools
  • Protégé, OntoEdit, WebODE, etc
  • Ontology Repositories
  • Sesame, OWLIM, etc
  • Ontology Reasoning
  • Pellet, Racer etc.
  • Ontology Integration Techniques
  • ontology mapping, alignment, merging

7
(Semantic) Web Services
  • Standard web services give us
  • loosely coupled, reusable interfaces
  • distributed and remotely accessible
  • programmatically accessible over standard
    internet protocols
  • alas only syntactic and thus open to confusion
  • Semantic web services give us
  • Additional semantic markup
  • Not just the syntax for interacting with the
    interface but the semantics to determine the
    meaning of it
  • Allow us to determine whether this is the right
    service

?
8
Using Semantic Web Services
  • Using web services in practice requires support
    for several tasks
  • Publication - to publish the existence of a
    service
  • Discovery - a mechanism to allow interested
    parties to find it
  • Selection - to choose the right service
    among the available
    ones
  • Composition - to be able to string several
    services together for
    more complex cases
  • Mediation - to mediate between otherwise
    incompatible services
  • Execution - to support the effective
    execution of services

9
OWL-S
  • OWL-S is an ontology to describe Web services
  • Based on existing Web Services Standards
  • rather than proposing another WS standard, OWL-S
    aims at enriching existing standards
  • OWL-S is grounded in WSDL and mapped into UDDI
  • But without an execution environment

Service
What it is
ServiceProfile
How to use it
How it works
Service Model
Service Grounding
10
WSMO
  • WSMO is a conceptual model for the core elements
    of Semantic Web Services
  • core elements Ontologies, Web Services, Goals,
    Mediators
  • language for semantic element description (WSML)
  • Uses ontologies as data model
  • every resource description is based on ontologies
  • every data element interchanged is an ontology
    instance
  • Has an execution framework WSMX

A Formal, Rule Based Modelling Language
WSMX Execution Environment
11
The Differences
  • OWL-S and WSMO use ontologies to represent
    semantics but are based on different logics
  • OWL-S is based on OWL / SWRL
  • OWL to represent the ontologies
  • SWRL to provides inference rules
  • FLOWS for process modeling
  • WSMO is based on WSML
  • Based on Description Logics and
  • Logic Programming
  • OWL-S is fully integrated into the existing web
    service standards (UDDI, WSDL, SOAP) but
    execution needs to be handled by the applications
    using them
  • WSMO operates like a middleware and has a
    sophisticated execution environment

12
So which ontologies do we need?
  • If we have to have a shared understanding to
    communicate between applications
  • We need to
  • Speak the same language
  • Or know how to translate
  • There are two basic types of ontologies
  • Domain ontologies which are specific to a
    particular application
  • Upper ontologies which are more general to allow
    us to categorise and communicate

13
Upper Ontologies
  • There are several upper ontologies
  • SUMO IEEE standardisation effort
    with 30kconcepts
  • OpenCyc general purpose (commercial)
  • DOLCE general purpose (logic oriented)
  • BFO general purpose (medical)
  • Wordnet linguistic ontology
  • OWL-S upper ontology for services

14
SUMO
  • The Suggested Upper Merged Ontology (SUMO) is the
    most prominent upper ontology
  • IT is being progressed by the IEEE Standard Upper
    Ontology (SUO) working group
  • It is an attempt to link categories and relations
    coming from different top-level ontologies to
    improve
  • interoperability,
  • communication and
  • search in the Semantic Web
  • The taxonomy is large and even features many
    domain concepts, such as Hotel or Organization.

15
Issues to discuss
  • So we have a lot of solutions
  • Semantic web technologies like RDF and OWL etc.
  • Semantic web services such as OWL-S and WSMO
  • Upper ontologies like SUMO etc
  • Standards including EDIFACT, ebXML etc.
  • Surely we should now be able to do business
    successfully?
  • So where is the problem?

16
(No Transcript)
17
Contentions
  • As a basis for discussion let us consider the
    following
  • Will upper ontologies really deliver what we want
    or do we need to live with a variety of
    ontologies?
  • We understand a lot about ontologies but we
    havent taught people how to build applications
  • We have created so much complexity that we cant
    deliver simple solutions any more
  • We havent yet convinced the industry that
    semantic technologies are robust and reliable to
    get them to develop products
Write a Comment
User Comments (0)
About PowerShow.com