CSci 586 Database Interoperability Course Presentation

1
CSci 586 Database InteroperabilityCourse
Presentation

• Presented by Srikanth Ranganamyna

2
About the Paper

• Title Ontology Languages for the Semantic Web
• Authors Asuncion Gomez-Perez and Oscar Corcho
• The authors analyze the most representative
  ontology languages created for web and compare
  them using common framework

3
Presentation Outlines

• Introduction
• Ontology Languages
• Comparison Criteria
• Conclusion

4
Introduction

• Ontologies are an essential part of many
  applications. They are used in-
• Agent Systems
• Knowledge Management systems
• E-commerce platforms
• To generate natural language
• Integrate intelligent information
• Provide semantic-based access to internet
• Ontologies triggered a major change in current
  web contents which lead to the third generation
  of web known as Semantic Web.

5
Introduction

• What is a Semantic Web ?
• It is defined as the conceptual structuring of
  the web in an explicit machine-readable way.
• What is an Ontology ?
• It is an explicit, machine readable specification
  of a shared conceptualization.
• What is required to make the semantic web
  functional ?
• We need languages to represent the semantic
  information that this web requires.

6
Introduction

• The goal of this paper is to help developers find
  a most suitable language for their
  representational needs and not to decide which
  language is best of the semantic web.

7
Ontology Languages
8
Ontology Languages

• Most of these ontology languages are based on
  XML, like -
• Ontology Exchange Language (XOL)
• Simple HTML Ontology Extension (SHOE)
• Ontology Markup Language (OML)
• Resource Description Framework (RDF)
• RDF Schema
• Some languages are built on top of RDF(S), like-
• Ontology Inference Language (OIL)
• DARPA Agent Markup Language (DAML) OIL

9
Ontology Exchange Language (XOL)

• Designed by US Bioinformatics Community
• It is used for the exchange of ontology
  definitions among a heterogeneous set of software
  systems in their domain.
• XOL is based on Ontolingua and OML.
• Ontolingua provides high expressiveness of
  knowledge (OKBC) and OML provides syntax based on
  XML.
• No tools are available to develop ontologies
  using XOL.

10
Simple HTML Ontology Extension (SHOE)

• Developed at University of Maryland.
• It is created as an extension of HTML,
  incorporating machine-readable semantic knowledge
  in HTML documents or other web documents.
• Enhances search mechanisms, knowledge gathering.
  This process consists of 3 phases-
• Define an Ontology
• Annotate HTML pages with ontological information
  to describe themselves and other pages
• Have an agent semantically retrieving the
  information from existing pages keep updating
  its information.

11
Ontology Markup Language (OML)

• Developed at University of Washington.
• It is partially based on SHOE.
• OML exists in 4 different levels, they are -
• OML Core- is related to the logical aspects of
  the language included by all other levels.
• Simple OML- maps directly to RDF(S).
• Abbreviated OML- includes the conceptual graph
  features.
• Standard OML- Most expressive version of OML.
• We discuss about Simple OML in this paper.
• No tools are available for authoring OML
  ontologies.

12
Resource Description Framework and RDF Schema

• Developed by W3C for describing the web
  resources.
• It allows the specification of the semantics of
  data based on XML in a standardized,
  interoperable manner.
• It consist of 3 objects-
• Resources
• Properties
• Statements

13
RDF and RDF Schema

• RDF data model does not define relationships
  between properties and resources but is provided
  by RDF Schema.
• RDFS offers primitives for defining knowledge
  models.
• RDF(S) is widely used as a representation format
  in many tools and projects like Amaya, Protege.

14
Ontology Interchange Language (OIL)

• Developed in OntoKnowlegde project.
• It allows semantic interoperability between web
  resources.
• Its syntax and semantics are based on existing
  proposals (OKBC, XOL and RDF(S)) providing
  modeling primitives commonly used in frame based
  approaches to Ontological Engineering (concepts,
  taxonomies of concepts, relations, etc).

15
Ontology Interchange Language (OIL)

• OIL as the following layers-
• Core OIL- groups the OIL primitives that have a
  direct mapping to RDF(S) primitives.
• Standard OIL- is the complete OIL model using
  more primitives than ones defined in RDF(S).
• Instance OIL- adds instances of concepts and
  roles to the previous model.
• Heavy OIL- is the layer for future extensions of
  OIL.
• OILEd, Protege and WebODE can be used to author
  OIL ontologies.

16
DARPA Agent Markup Language OIL(DAML OIL)

• Developed by the joint committee from US and
  European Union in the context of DAML, a DARPA
  project for allowing semantic interoperability in
  XML.
• DAML and OIL shares the same objective.
• DAMLOIL replaces the initial specification,
  called DAML-ONT.
• OntoEdit, Protege, WebODE can be used to author
  DAMLOIL ontologies.

17
Comparison Criteria for Ontology Languages

• This paper uses the same set of criteria to
  compare different ontology languages , based on
  different representational formalisms.
• Ontology knowledge can be specified using 5
  components, they are-
• Concepts (usually organized by taxonomies)
• Relations
• Functions
• Axioms
• Instances.

18
Concepts

• Concepts could be anything from being abstract,
  concrete, simple, composite, real or fictitious
  and therefore could be a description of the task,
  function, strategy, reasoning process, etc.
• Concepts are also known as Classes (XOL, RDF(S),
  DAMLOIL), Objects (OML) or Categories (SHOE).

19
Concepts

• Several questions are asked before evaluating
  concepts, they are-
• Is it possible to define partitions (sets of
  disjoint concepts) ?
• Is it possible to define the documentation of the
  concept ?
• Does the language provide mechanisms to define
  the concept attributes ?
• Attributes are also known as slots (XOL),
  functions ( OML), or properties (RDF(S))

20
Concepts

• Attributes are of the following types in
  concepts-
• Instance attributes
• Class attributes
• Local attributes
• Global attributes
• Instance and Class attributes are commonly used
  in concept descriptions. Local and Global
  attributes requirement depends on the
  applications representation needs.

21
Concepts

• Does the language provide the following
  predefined facets for attributes at a class ?
• Default Slot Value
• Type
• Cardinality constraints
• Slot Documentation

22
Taxonomies

• Taxonomies are used to organize ontological
  knowledge using generalization and specialization
  relationships through which simple and multiple
  inheritance could be applied.
• We analyze whether the following primitives are
  predefined in the languages
• Subclass of
• Disjoint decomposition
• Exhaustive subclass decomposition
• Not subclass of

23
Relations and Functions

• Relations are an interaction between concepts of
  the domain and attributes.
• Functions are a special kind of relations where
  the value of the last argument is unique for a
  list of values of the n-1 preceding arguments.
• The following questions are asked before
  evaluating
• Is it possible to define arbitrary n-ary
  relations or functions ?

24
Relations and Functions

• Might the type of the arguments be constrained ?
• Is it possible to define any kind of integrity
  constraints to check the correctness of arguments
  ?
• Is it possible to define operational definitions
  to infer values of arguments with procedure,
  formulas or rules ?

25
Axioms

• Axioms model sentences that are always true.
• They are used for constraining information,
  verifying correctness, or deducing new
  information.
• Axioms are also known as assertions (OML)
• We focus on first on following characteristics
• Is it possible axioms in first order logic,
  second order logic ? ( I order pgtp, II order
  g(p) gt h(p) for all f(p))
• Can we build axioms as independent elements in
  ontology ? (Independent Axioms)
• Can we embed axioms inside the definition of
  other elements ?

26
Instances

• Instances represent elements in domain attached
  to a specific concept.
• The following questions are asked before
  evaluating a language on this criteria
• Is it possible to define instances of concepts ?
  (eg Peter is an instance of class buyer).
• Is it possible to define instances of relations ?
• Does the language provide special mechanisms to
  define claims ? (eg John says that Peter has
  brought computer2)
• Information in distributed resources on semantic
  web is described using instances of concepts and
  relations.

27
Results for Concepts
28
Results for Taxonomies
29
Results for Relations and Functions
30
Results for Axioms
31
Results for Instances
32
Experiment An Ontology for an E-Commerce Platform
33
Conclusion

• The ontology language should be selected based on
  the representational needs of an application.
• Ontology languages make the semantic web
  functional because semantic web needs a language
  which can represent semantic information.
• DAMLOIL Ontology language is like a superset of
  other ontology languages in terms of the features
  offered for developers.

34
References

• T. Berners-Lee and M. Fischetti, Weaving the Web
  The Original Design and Ultimate Destiny of the
  World Wide Web by its Inventor, Harper, San
  Francisco, 1999.
• R. Karp,V. Chaudhri, and J. Thomere, XOL An
  XML-Based Ontology Exchange Language (version
  0.4), Aug. 1999, www.ai.sri.com/pkarp/xol
  (current Jan. 2002).
• S. Luke and J. Heflin, SHOE 1.01 Proposed
  Specification, SHOE Project, Feb. 2000.
  www.cs.umd.edu/projects/plus/SHOE/spec1.01.htm
  (current Jan. 2002).

35
Questions and Sugestions