A Brief Introduction to Ontologies and Relative Web Development for them

1
A Brief Introduction to Ontologies and Relative
Web Development for them

• Shui-Lung Chuang
• August 30, 2001

2
Example To Solve a Block-world Problem

(1)
Solution Process Move( A, Table ) Move( B,
A ) Move( C, B )
C
A
B
C
B
A
Goal
Initial State
(2)
C on Table B on Table A on B
Block-world problem solver if ( A on Table )
then . else if ( B ) then else if (
C ) then .
A
B
C
Move( A, Table ) Move( B, A ) Move( C, B )
Initial State
3
Example To Solve a Block-world Problem (cont.)

(3) Conceptualization
Thing
Relation
Entity
Table
Block
Hand
Binary
Unary
handEmpty
table X
block A
hand A
on
above
clear
block C
block B
on(A,C)
Axiom above(X,Z)-on(X,Y) and on(Y,Z)
A General Problem Solver

4
Reuse and Sharing

• Reuse
• To build new applications assembling components
  already built
• Sharing
• To use the same resources in different
  applications
• Reusable components of knowledge-based system
• Ontology
• generic modeling of domain knowledge
• Problem-solving methods
• domain-independent reasoning or planning process

5
Evolution of Programming Paradigms

Time immemorial machine code/assembly language
Make computational process more explicit more
modularized more conceptualized more
naturally-represented
Abstract Data Type (ADT) Structure Programming
Object-Oriented Paradigm

Component-based
Pattern-based
Framework-based
6
Object-Oriented Paradigm
Object
Conceptualization of the World

Real
Animal
Plant
class Human properties string name
sex male, female int age not lt
0 actions walk() eat()

Instantiated World
john Human( John, male, 25 ) john.walk() and
john.eat() // John ????
7
What is an Ontology?

• An ontology is an explicit specification of
  a conceptualization. The term is borrowed from
  philosophy, where an ontology is a systematic
  account of Existence. For knowledge-based
  systems, what exists is exactly that which can
  be represented. When the knowledge of domain is
  represented in a declarative formalism, the set
  of objects that can be represented is called the
  universe of discourse. This set of objects, and
  the describable relationships among them, are
  reflected in the representational vocabulary with
  which a knowledge-based program represents
  knowledge. Thus, we can describe the ontology of
  a program by defining a set of representational
  terms. In such an ontology, definitions
  associate the names of entities in the universe
  of discourse (e.g., classes, relations,
  functions, or other objects) with human-readable
  text describing what the names are meant to
  denote, and formal axioms that constrain the
  interpretation and well-formed use of these
  terms.
• Thomas R. Gruber, 1993

8
What is an Ontology? (cont.)

• Philosophy
• The subject or theory of existence
• What is existence? What properties can explain
  existence?
• Epistemology
• Knowledge and knowing
• AI community
• An explicit representation of a conceptualization
• KB community
• A theory of vocabulary/concepts used as building
  artificial systems

9
What is an Ontology? (cont.)

• To be more precise, an ontology
• contains a taxonomy of important concepts in a
  domain,
• describes crucial properties of each concept
  through an attribute-value mechanism,
• has further relations between concepts described
  in additional logical sentences,
• has individuals assigned to one or more concepts.
• Some characteristics of ontologies
• Human-readable
• Declarative rather than procedural description
• Machine-interpretable
• represented by some formal language, e.g., KIF

10
Why develop an Ontology?

• Some reasons to develop an ontology are
• To share common understanding of the structure of
  information among people or software agents
• To enable reuse of domain knowledge
• To make domain assumptions explicit
• To separate domain knowledge from the operational
  knowledge
• To analyze domain knowledge
• you name it!

11
What is in an Ontology?

• Concepts (or classes) are organized in taxonomies
• Properties (or slots, or roles)
• Relations and functions
• Instances
• Axioms
• Sentences which are always true

Thing
Relation
Entity
handEmpty
Binary
Block
Hand
Table
Unary
table X
block A
hand A
block C
on
above
clear
block B
on(A,C)
Axiom above(X,Z) - on(X,Z)
above(X,Z)- above(X,Y) above(Y,Z)
12
Example Yahoo! and Thesaurus

• Yahoo!-like Web directories
• Thesaurus
• Relations
• BT
• NT
• Synonym
• RT

http//111/111.html\ http//222/222.htm http//333
/333.html
13
Example WordNet

• A lexical database (a taxonomy, no concepts and
  axioms)
• Correspondence between terms and meanings
• Categories
• Nouns organized in hierarchies by hypernym
  and hyponym
• Verbs Implication relationships
• Adj. and Adv. N-dimensional hyperspaces
• Example
• Board -gt board,plank
• Board -gt board,committee
• Board -gt board -gt a persons meals,
  provided regularly for money

14
Example CYC

• CYC Project To create a general ontology for
  commonsense knowledge.

15
Ontology Engineering

• The principled design, maintenance, and
  application of ontologies
• Current research directions
• Make ontologies sharable by developing common
  formalisms and tools
• formal language, logic,
• Develop the content of ontologies
• techniques (NLP, data mining, machine learning)
  for complementing human efforts
• annotation tools
• Compare, gather, translate, and compose different
  ontologies
• intelligent information integration

16
Web Development and Ontologies

• Intelligent Agents
• E-Commerce
• Query expansion, parametric search, machine-based
  communication between buyer and seller,
• Knowledgeable Portals (Semantic Web)

17
Web Standards

• Various standards make WWW possible and guarantee
  interoperability at its various levels.
• TCP/IP
• HTTP/FTP/TELNET
• HTML
• XML/XML Schema
• RDF/RDF Schema
• OIL

Declarative Language, e.g., OIL
RDF
XHTML
HTML
XML
18
eXtensible Markup Language (XML)

• XML provides a standard format (syntax) to
  represent any resources on the Web.
• Well-formed
• Valid

ltclass-defgt ltclass nameplant/gt
ltsubclass-ofgt ltNOTgtltclass nameanimal/gtlt/NOT
gt lt/subclass-ofgt lt//class-defgt ltclass-defgt
ltclass nametree/gt ltsubclass-ofgt ltclass
nameplant/gt lt/subclass-ofgt lt/class-defgt ltclas
s-defgt ltclass namebranch/gt
ltslot-constraintgt ltslot nameis-part-of/gt
lthas-valuegt ltclass nametree/gt lt/has-valuegt
lt/slot-constraintgt lt/class-defgt
Example
not subclass-of
plant
animal
subclass-of
ltclass-defgt ltnamegtbranchlt/namegt
ltslot-constraintgt ltnamegtis-part-oflt/namegt
lthas-valuegttreelt/has-valuegt lt/slot-constraintgt lt
/class-defgt
is-part-of
tree
branch
19
Resource Description Framework (RDF)

• RDF is for modeling the meta-data about the
  resources of the Web.

Resources are named by URIs. Anything can have a
URI.
? object-attribute-value model
attribute
object
value
? Example
http//purl.org/DC/Creator
http//rama.cpe.fr/index.html
http//rama.cpe.fr/index.html
http//purl.org/DC/Title
lt?xml version1.0 encodingUTF-8 ?gt ltrdfRDF
xmlnsrdfhttp//www.w3.org//22-rdf.syntax-ns
xmlnsdchttp//purl.org/DC/gt
ltrdfDescription abouthttp//rama.cpe.fr/index.h
tmlgt ltdcCreator rdfresourcemailtoam_at_cpe.
fr/gt ltdcTitlegt Index of my web site
lt/dcTitlegt lt/rdfDescriptiongt lt/rdfRDFgt
Index of my web site
20
More than RDF

• Ontology Inference Layer (OIL)

Description Logics Formal Semantics Reasoning
Support
Frame-based systems Epistemological
Modeling Primitives
OIL
Web Languages XML- and RDF-based syntax
21
Conclusion

• An ontology is a formal, explicit specification
  of a shared conceptualization.
• New generation of the Web provides a test bed for
  applying ontologies.
• Standardized representational syntax XML, RDF,
• Semantic Web
• Intelligent agents
• .

Concepts, properties, relations, instances, and
axioms are explicitly defined
Consensual knowledge
Machine-readable
Abstract model of some phenomenon in the world
22
References

• Ontology
• Thomas R. Gruber, A Translation Approach to
  Portable Ontology Specifications, Technical
  Report KSL 92-71, Knowledge Systems Laboratory,
  Stanford University.
• Asuncion Gomez-Perez, Tutorial on Ontological
  Engineering, IJCAI99.
• Natalya Fridman Noy and Carole D. Hafner, The
  State of the Art in Ontology Design, AI Magazine,
  18(3), 53-74, 1997.
• Riichiro Mizoguchi and Mitsuru Ikeda, Towards
  Ontology Engineering, Technical Report
  AI-TR-96-1, I.S.I.R. Osaka University.
• Semantic Web, XML, RDF, and OIL
• Tim Berners-Lee, Semantic Web Road Map,
  http//www.w3.org/DesignIssues/Semantic.html
• Stefan Decker, etc. The Semantic Web The Roles
  of XML and RDF, IEEE Network Computing,
  Sep.-Oct., 2000.
• Pierre-Antoine Champin, RDF Tutorial, 2000.
  http//bat710.univ-lyon1.fr/champin/rdf-tutorial/
  
• D. Fensel, etc, OIL in a Nutshell, Proceedings of
  the European Knowledge Acquisition Conference
  (EKAW-2000), R. Dieng et al. (eds.), Lecture
  Notes on Artificial Intelligences, LNAI, Springer
  Verlog, October 2000.
• Some good sites about ontology
• Mizoguchi Lab Home Page, ISIR, Osaka Univ.
  http//www.ei.sanken.osaka-u.ac.jp/
• Stanford Knowledge Systems Laboratory
  http//www-ksl.stanford.edu/
• ANSI Ad Hoc Group on Ontology Standards
  http//www-ksl.stanford.edu/onto-std/