Transcript Knowledge Representation I (pptx

Knowledge Representation
Part I
Ontology
Jan Pettersen Nytun
Knowledge Representation Part I, JPN, UiA
1
S
P
O
Outline
• Knowledge
• Reasoning / logical Consequence
• Ontology
– Ontology in philosophy
– Ontology in computer science
– Different types of ontologies
• Levels of ontological precision
Knowledge Representation Part I, JPN, UiA
2
facts/understanding
about a particular subject
a symbol or thing which represents
something else (refers to, stands for)
is
is
Knowledge Representation
AI require
when to use
computer-understandable form
when we can not use the “original”, like
things in the natural world or concepts
Knowledge Representation Part I, JPN, UiA
3
S
P
O
From Wikipedia, the free encyclopedia
(Knowledge representation and reasoning)
Knowledge Representation (KR) is an
area of artificial intelligence research aimed
at representing knowledge in symbols to
facilitate inferencing from those knowledge
elements, creating new elements of
knowledge.
Knowledge Representation Part I, JPN, UiA
4
S
P
O
Knowledge Base
• A database for knowledge management
• It provides means for information to be:
– Collected
– Organized
– Shared, searched and utilized (new information
may be inferred)
Knowledge Representation Part I, JPN, UiA
5
S
P
O
Knowledge Engineering
• Get knowledge about some subject and
represent it in a computable form for
some purpose.
• The knowledge engineer tells the system
what is true.
Knowledge Representation Part I, JPN, UiA
6
S
P
O
Outline
• Knowledge
• Reasoning / logical Consequence
• Ontology
– Ontology in philosophy
– Ontology in computer science
– Different types of ontologies
• Levels of ontological precision
Knowledge Representation Part I, JPN, UiA
7
S
P
O
Asserted and Inferred Statements
The system knows how to infer new facts and solutions – the
user may form questions and then the system gives answers.
Knowledge Base
Asserted
Statements
Asserted
Statements
Inferred
Statements
Entailment
Inferred statements comes as a logical
consequence of the asserted statements
and logical rules
Knowledge Representation Part I, JPN, UiA
Inferred
Statements
8
S
P
O
Entailment (Logical Consequence)
Example: Family Information
• Identify “something” as being Person:
Person(Ola), Person(Kari), Person(Marie),
Person(Jan), …
• Gender of person:
Female(Kari), Male(Ola), Female(Marie),
Male(Jan), …
• Who is parent to a person:
Parent(Ola, Marie), Parent(Kari, Marie), …
Knowledge Base
Asserted Statements:
Inferred Statements
Person(Ola), Person(Kari),
Person(Marie),Person(Jan),
Female(Kari), …
Knowledge Representation Part I, JPN, UiA
9
S
P
O
Example: Family Information … Continues
Given the right logical rules, then family relations can be derived:
• Parent(x, y) and Female(x)  Mother(x, y)
• ??  Daughter (x, y)
• ??  Brother(x, y)
Knowledge Base
Asserted Statements:
Inferred Statements:
Person(Ola), Person(Kari),
Person(Marie),Person(Jan),
Female(Kari), Male(Ola),
Female(Marie), Male(Jan),
Parent(Ola, Marie),
Parent(Kari, Marie), …
Mother(Kari, Marie), …
Knowledge Representation Part I, JPN, UiA
10
S
P
O
Complex relations:
Knowledge Representation Part I, JPN, UiA
11
S
P
O
Outline
• Knowledge
• Reasoning / logical Consequence
• Ontology
– Ontology in philosophy
– Ontology in computer science
– Different types of ontologies
• Levels of ontological precision
Knowledge Representation Part I, JPN, UiA
12
S
P
O
What is an Ontology
in Regard to Philosophy?
From Wikipedia, the free encyclopedia
13
S
P
O
What is an Ontology in Regard to Philosophy? Continues…
Smith [1] the essence of
ontology:
“provide a definitive and
exhaustive classification
of entities in all spheres
of being.”
14
S
P
O
What is an Ontology
in Computer Science?
Knowledge represented in a formal way:
- a hierarchy of concepts within a domain,
- a shared vocabulary to denote the types,
- properties and interrelationships of those
concepts.
15
S
P
O
What is an Ontology
in Computer Science? … Continues
An ontology is a specification of a conceptualization that is
designed for reuse across multiple applications and
implementations. …a specification of a conceptualization
is a written, formal description of a set of concepts and
relationships in a domain of interest.
Peter Karp (2000) Bioinformatics 16:269
16
S
P
O
Ontology vs Knowledge Base"
“The Artificial-Intelligence literature contains many
definitions of an ontology; many of these contradict
one another. … An ontology together with a set of
individual instances of classes constitutes a
knowledge base. In reality, there is a fine line
where the ontology ends and the knowledge base
begins.”
[http://protege.stanford.edu/publications/ontology_development/ontology101-noy-mcguinness.html]
Knowledge Representation Part I, JPN, UiA
17
S
P
O
Not All Would Agree On The
Following:
• “An ontology is, very roughly, a formal representation of
a domain of knowledge. It is an abstract entity: it defines
the vocabulary for a domain and the relations between
concepts, but an ontology says nothing about how that
knowledge is stored (as physical file, in a database, or in
some other form), or indeed how the knowledge can be
accessed.
• A knowledge base is a physical artifact: it is a database, a
repository of information that can be accessed and
manipulated in some predefined fashion. The knowledge
in a knowledge base can be said to be modeled according
to an ontology.”
[http://answers.semanticweb.com/questions/21500/what-is-the-difference-between-knowledge-base-and-ontology]
Knowledge Representation Part I, JPN, UiA
18
Types of Ontologies
[Ref. Medical Informatics: Knowledge Management and Data Mining in Biomedicine]:
From Wikipedia, the free encyclopedia:
In computer science and information science, an ontology is…
a practical application of philosophical ontology.
19
S
P
O
Types of Ontologies… Continues
An upper ontology - also called top-level
ontology or foundation ontology - describes the
most general concepts that are the same across
all knowledge domains (e.g., Entity).
Knowledge Representation Part I, JPN, UiA
20
S
P
O
Types of Ontologies… Continues
[Ref. Medical Informatics: Knowledge Management and Data Mining in Biomedicine]:
General ontologies represent
knowledge at an intermediate level of
detail independently of a specific task…
theories of time and space, for example...
Knowledge Representation Part I, JPN, UiA
21
S
P
O
Types of Ontologies… Continues
[Ref. Medical Informatics: Knowledge Management and Data Mining in Biomedicine]:
Domain ontologies represent knowledge about a
particular part of the world, such as medicine, and
should reflect the underlying reality through a
theory of the domain represented.
Knowledge Representation Part I, JPN, UiA
22
S
P
O
Types of Ontologies… Continues
[Ref. Medical Informatics: Knowledge Management and Data Mining in Biomedicine]:
…ontologies designed for specific tasks are called
application ontologies. Conversely, reference
ontologies are developed independently of any
particular purpose…
Knowledge Representation Part I, JPN, UiA
23
Descriptive Ontology for Linguistic and Cognitive Engineering
S
P
O
Knowledge Representation Part I, JPN, UiA
24
S
P
O
Outline
• Knowledge
• Reasoning / logical Consequence
• Ontology
– Ontology in philosophy
– Ontology in computer science
– Different types of ontologies
• Levels of ontological precision
Knowledge Representation Part I, JPN, UiA
25
S
P
O
Catalog: A list of things.
Knowledge Representation Part I, JPN, UiA
26
S
P
O
From Wikipidia:
A Glossary, also known as a vocabulary,… is an alphabetical list of terms in a
particular domain of knowledge with the definitions for those terms.
S
P
O
A Taxonomy – also called a class hierarchy - organizes its data into
categories and subcategories.
S
P
O
In general usage, a thesaurus is a reference work that lists words
grouped together according to similarity of meaning (containing synonyms and
sometimes antonyms).
From Wikipidia:
S
P
O
From Wikipidia:
A database schema …is a structure described in a formal
language… and refers to the organization of data as a blueprint of how a database is
constructed (e.g., database tables for Relational Databases).
In mathematics, an axiomatic system is any set of
axioms from which some or all axioms can be used in conjunction to
logically derive theorems.
A mathematical theory consists of an axiomatic system and all its
derived theorems.
From Wikipidia:
S
P
O
Ontology Engineering
as a Discipline
Example of Process
Decide Scope
Reuse?
Enumerate Terms
Studies the methods and
methodologies for building
ontologies.
Define Classes
Define Properties
Define Constraints
Knowledge Representation Part I, JPN, UiA
Create Instances 32
S
P
O
References
[1] Book: David Poole and Alan Mackworth, Artificial Intelligence: Foundations of
Computational Agents, Cambridge University Press, 2010, http://artint.info/
Sowa, John F. (2000) Knowledge Representation: Logical, Philosophical, and
Computational Foundations, Brooks/Cole Publishing Co., Pacific Grove, CA.
Artificial Intelligence: Structures and Strategies for Complex Problem Solving (AddisonWesley), George F. Luger
Smith Barry. Accessed 24th of March, 2013, Ontology: Philosophical and Computational.
http: //ontology.buffalo.edu/smith/articles/ontologies.htm
Quine WVO. On What There Is. Review of Metaphysics 1948;p. 21–38.
Knowledge Representation Part I, JPN, UiA
33