Loading...

Ontological Engineering approach of developing Ontology of Information Science

©2015 Textbook 298 Pages

Summary

Ontology has been a subject of many studies carried out in artificial intelligence (AI) and information system communities. Ontology has become an important component of the semantic web, covering a variety of knowledge domains. Although building domain ontologies still remains a big challenge with regard to its designing and implementation, there are still many areas that need to create ontologies. Information Science (IS) is one of these areas that need a unified ontology model to facilitate information access among the heterogeneous data resources and share a common understanding of the domain knowledge. Recently, the development of domain ontologies has become increasingly important for knowledge level interoperation and information integration.
They provide functional features for AI and knowledge representation. Domain Ontology is a central foundation of growth for the semantic web that provides a general knowledge for correspondence and communication among heterogeneous systems. Particularly with a rise of ontology in the artificial intelligence (AI) domain, it can be seen as an almost inevitable development in computer science and AI in general.

Excerpt

Table Of Contents


IV
Acknowledgments
First of all a special thank to my Lord Almighty Allah. We always keep the deficit in the
description of words of thanks, especially to people who provide with tenderness without
borders, the lines of thanks are always extremely difficult to formulate.
I would like to thank Professor Zhongyu (Joan) Lu, for her support, encouragement, and
guidance from the initial to the final level, enabling me to develop an understanding of
the subject, and her excellent feedback on the related paper and articles produced
during the research project. Her encouragement has a big impact on my learning in
finding pieces of knowledge and solving challenges.
I am heartily thankful to Julie Wilkinson and Violeta Holmes for their guidance and
advice during these years.
Also, great thanks to all the staff of the Computing and Engineering School at the
University of Huddersfield who inspired me to do my research. A big thank to everyone
who participated in Ontocop's project.
Also, all the people were at my side; my sincere and faithful husband for his great and
magnificent support, and the three essences of my life: Khalifa, Tasnim and Noor.
A great thanks also to all who gave me a helping hand to reach the end of my journey.

V
Abstract
Ontology has been a subject of many studies carried out in artificial intelligence (AI) and
information system communities. Ontology has become an important component of the
semantic web, covering a variety of knowledge domains. Although building domain
ontologies still remains a big challenge with regard to its designing and implementation,
there are still many areas that need to create ontologies. Information Science (IS) is one
of these areas that need a unified ontology model to facilitate information access among
the heterogeneous data resources and share a common understanding of the domain
knowledge. The objective of this study is to develop a generic model of ontology that
serves as a foundation of knowledge modelling for applications and aggregation with
other ontologies to facilitate information exchanging between different systems. This
model will be a metadata for a knowledge base system to be used in different purposes
of interest, such as education applications to support educational needs for teachers and
students and information system developers, and enhancing the index tool in libraries to
facilitate access to information collections. This thesis describes the process of modelling
the domain knowledge of Information Science IS.
The building process of the ontology of Information Science (OIS) is preceded by
developing taxonomies and thesauruses of IS. This research adopts the Methontology to
develop ontology of Information Science OIS. This choice of method relies on the
research motivations and aims, with analysis of some development ontology
methodologies and IEEE 1074-2006 standards for developing software project life cycle
processes as criteria. The methodology mainly consisted of; specification,
conceptualization, formalization, implementation, maintenance and evaluation. The
knowledge model was formalized using Protégé to generate the ontology code. During
the development process the model has been designed and evaluated.
This research presents the following contributions to the present state of the art on
ontology construction;
-
The main achievement of the study is in constructing a new model of Information
Science ontology OIS. The OIS ontology is a generic model that contains only the
key objects and associated attributes with relationships. The model has defined
706 concepts which will be widely used in Information Science applications. It
provides the standard definitions for domain terms used in annotation databases
for the domain terms, and avoids the consistency problems caused by various
ontologies which will have the potential of development by different groups and
institutions in the IS domain area.

VI
-
It provides a framework for analyzing the IS knowledge to obtain a classification
based on facet classification. The ontology modelling approach is based on top-
down and bottom­up. The top-down begins with an abstract of the domain view.
While the bottom-up method starts with description of the domain to gain a
hierarchal taxonomy.
-
Designing Ontocop system a novel method presented to support the developing
process as specific virtual community of IS. The Ontocop consists of a number of
experts in the subject area around the world. Their feedback and assessment
improve the ontology development during the creating process.
The findings of the research revealed that overall feedback from the IS community has
been positive and that the model met the ontology quality criteria. It was appropriate to
provide consistency and clear understanding of the subject area. OIS ontology unifies
information science, which is composed of library science, computer science and archival
science, by creating the theoretical base useful for further practical systems. Developing
ontology of information science (OIS) is not an easy task, due to the complex nature of
the field. It needs to be integrated with other ontologies such as social science, cognitive
science, philosophy, law management and mathematics, to provide a basic knowledge
for the semantic web and also to leverage information retrieval.

VII
Publications
1
Sawsaa, A. & Lu, J (2009). A Generic Model of Knowledge Mapping Through
Virtual Communities of Practice in Information Science (IS). Conference
proceeding. World Congress in Computer Science, Computer Engineering,
and Applied Computing, Las Vegas, Nevada.12-15 July 2009.
2
Sawsaa, A. & Lu, J (2010). Ontocop: A virtual community of practice to
create ontology of Information science (IS). Conference proceeding. World
Congress in Computer Science, Computer Engineering, and Applied
Computing, Las Vegas, Nevada 12-15 July 2010.
3
Sawsaa, A. & Lu, J (2010). Ontology of Information Science (IS) based on
OWL conference proceeding. The International Arab Conference on
Information Technology (ACIT2010).
4
Sawsaa, A. (2010). A virtual community. The 3th Scientific Research
Symposium for Libyan Students in UK Universities. Sheffield Hallam
University, 12th June 2010.
5
Sawsaa, A. & Lu, J. (2011) `Extracting Information Science concepts based
on Jape Regular Expression'. In: WORLDCOMP'11 - The 2011 World
Congress in Computer Science, Computer Engineering, and Applied
Computing, 18-21 July 2011, Las Vegas, Nevada, USA
6
Sawsaa, A. & Lu, J. (2011) `Virtual Community of Practice Ontocop: Towards
a New Model of Information Science Ontology (OIS)' International Journal of
Information Retrieval Research, 1 (2), pp. 55-78. ISSN 2155-6377
7
Sawsaa, A., ZHAOZONG, M. & LU, J. (2012) Using an Application of Mobile
and Wireless Technology in Arabic Learning System. IN LU, Z. J. (Ed.
Learning with Mobile Technologies, Handheld Devices and Smart Phones:
Innovative Methods. USA, IGI Global. pp. 171-186. ISBN978-1-4666-0936-5
8
Sawsaa, A. & Lu, J. (2012) Developing a Domain Ontology of Information
Science (OIS). IN SHONIREGUN, C. A. & AKMAYEVA, G. A. (Eds.)
International Conference on Information Society (i-Society 2012) June 25-
28, 2012,. London, UK, i-Society 2012 Technical Co-Sponsored by IEEE
UK/RI Computer Chapter. pp 462-467.
9
Sawsaa, A. & Lu, J. (2012) Extracting Information Science Concepts based
on JAPE Regular Expression.
International Journal of advanced Computer
Science (IJEC) , in press

VIII
10
Sawsaa, A. & Lu, J. (2012) Building an advance domain ontology model of
Information Science (OIS).
Journal of the American Society for Information
Science and Technology (JASIST), in press
11
Sawsaa, A. & Lu, J. (2012) Building Information Science (OIS) Ontology with
Methondology and Protégé. Journal of Internet Technology and Secured
Transactions (JITST) 2 (1/2), ISSN 2046-3723. In press.
12
Sawsaa, A. & Lu, J. (2013) Extracting Occupational Therapist concepts to
develop domain ontology. The Seventh International Conference on Digital
Society (ICDS 2013)
February 24 - March 1, 2013 - Nice, France.
Posters:
1
Sawsaa, A. and Lu, J. (2011) `ONTOCOP: Virtual Community Of Practice to
build Ontology of Information Science'. University of Huddersfield 2009
2
Sawsaa, A. and Lu, J. (2009) A Generic Model of Knowledge Mapping
Through Virtual Communities of Practice in Information Science (IS)

IX
Table of contents
ACKNOWLEDGMENTS ... IV
ABSTRACT ... V
PUBLICATIONS ... VII
TABLE OF CONTENTS ... IX
LIST OF FIGURES ... XVII
LIST OF TABLES ... XXI
ACRONYMS ... XXIII
PART 1: FUNDAMENTAL ISSUES ... 1
1
CHAPTER 1: INTRODUCTION ... 2
1.1.
P
ROBLEM
I
DENTIFICATION
... 3
1.2.
A
IMS AND
O
BJECTIVES
... 4
1.3.
M
ETHODOLOGY AND
I
MPLEMENTATION
... 5
1.4.
C
ONTRIBUTIONS
... 6
1.5.
M
OTIVATION OF STUDY
... 8
1.6.
T
HESIS ORGANIZATION
... 9
2
CHAPTER 2: RESEARCH BACKGROUND ... 12
2.1
O
NTOLOGY
O
VERVIEW
... 12
2.1.1
Historical and philosophical perspective of the ontology ... 13
2.1.1.1
Definition of ontology ... 14
2.1.2
Ontology Theoretic ... 16
2.1.2.1
Category Theory ... 16
2.1.2.2
Mereotopolgy Theory ... 17
2.1.3
Referencing and meaning in the ontology ... 19
2.1.4
Ontology spectrum ... 20
2.1.4.1
Thesaurus ... 20

X
2.1.4.2
Taxonomy ... 21
2.1.5
Approaches for modelling ontology: ... 24
2.1.5.1
Top-down approach ... 24
2.1.5.2
Bottom­up approach ... 24
2.1.5.3
Middle out approach ... 25
2.1.6
Structure of ontology... 26
2.1.7
Ontology Categorization ... 27
2.1.7.1
Informal ontology ... 29
2.1.7.2
Formal ontology... 29
2.1.7.3
Domain ontology ... 29
2.1.8
Related Research ... 30
2.1.9
Designing Criteria for ontology ... 35
2.1.10
Ontology evaluation approaches ... 36
2.1.11
Ontology Engineering Methodologies ... 39
2.1.11.1
CYC Method ... 40
2.1.11.2
Uschold & King Method. ... 40
2.1.11.3
Gruninger& Fox Method. ... 41
2.1.11.4
SENSUS Methodology ... 42
2.1.11.5
Methontology ... 42
2.1.11.6
Comparison of Methodology ... 46
2.1.11.7
Evaluation of ontology methodologies ... 49
2.1.12
Techniques Involved ... 50
2.1.12.1
Ontology languages ... 50
2.1.12.2
Resource Description Framework RDF, and RDFs ... 51
2.1.12.3
Web Ontology Language (OWL) ... 54

XI
2.1.12.4
Comparison of ontology languages ... 57
2.1.12.5
Ontology Tools ... 58
2.1.12.6
Ontologua server ... 59
2.1.12.7
OntoSaurus ... 59
2.1.12.8
WebOnto ... 59
2.1.12.9
OilEd: ... 59
2.1.12.10
Cmap tools ... 60
2.1.12.11
Protégé ... 60
2.1.12.12
Web Protégé ... 60
2.1.12.13
General Architecture for text engineering( GATE) ... 60
2.1.12.14
Comparison of ontology tools ... 61
2.2
I
NFORMATION
S
CIENCE
(IS) ... 63
2.2.1
Overview... 63
2.2.2
Definitions ... 63
2.2.3
Relationship of information science with other sciences... 65
2.2.4
Information Science Taxonomy ... 67
2.2.4.1
Universal decimal classification (UDC) ... 69
2.2.4.2
Library of Congress Classification (LCC) ... 69
2.2.4.3
Colon Classification Scheme (CCS) ... 69
2.2.4.4
The advantages of Facet analysis system (FAS) ... 70
2.2.4.5
Classification Research group (CRG) ... 72
2.2.5
Why Information Science Taxonomy: ... 72
2.3
K
NOWLEDGE MANAGEMENT
(KM)
AND
V
IRTUAL COMMUNITIES OF
P
RACTICE
(VC
OPS
). ... 74
2.3.1
The Main components of knowledge management. ... 74
2.3.1.1
Data ... 74

XII
2.3.1.2
Information ... 75
2.3.1.3
Knowledge ... 75
2.3.2
Knowledge Management ... 76
2.3.3
Knowledge Engineering (KE)... 78
2.3.4
Knowledge Representation (KR) ... 79
2.3.5
Virtual Communities of practice (VCops). ... 81
2.3.6
Communities of Practice (Cops) ... 81
2.3.7
Virtual communities of practice (VCops) ... 83
2.3.8
Summary ... 88
PART 2: METHODOLOGY OF CREATING ONTOLOGY OF INFORMATION SCIENCE (OIS) ... 89
3
CHAPTER 3: METHODOLOGY EMPLOYED ... 90
3.1
T
HEORETICAL
A
PPROACHES
... 90
3.1.1
Taxonomy of OIS ontology approach ... 91
3.2
T
HE METHODOLOGY TO BE ADOPTED
... 94
3.3
T
ECHNIQUES AND
T
OOLS TO BE EMPLOYED
... 94
3.4
E
STABLISHING THE ONTOLOGY MODEL
... 95
3.4.1
Conceptual aspect ... 95
3.4.2
Computational aspect ... 96
3.5
I
NTRODUCING
OIS
DESIGN METHODOLOGY
... 96
3.5.1
Designing ontology model ... 96
3.5.2
Designing ontocop website tool ... 100
3.5.3
Summary ... 100
PART 3: IMPLEMENTATION ... 101
4
CHAPTER 4: MODELLING DESIGN OF OIS ONTOLOGY ... 102
4.1
B
UILDING
C
ONCEPTUAL
M
ODEL
... 102

XIII
4.1.1
Specifications ... 102
4.1.1.1
Identifying the purpose and the scope ... 102
4.1.1.2
Knowledge acquisition ... 104
4.1.2
Conceptualisation of IS entities ontology ... 109
-
Identification of concepts and relations ... 109
4.1.2.1
Building Glossary of terms of IS ... 110
4.1.2.2
Building Concepts taxonomy ... 110
4.1.2.3
Building ad hoc binary relation: ... 116
4.1.2.4
Build the concept dictionary ... 116
4.1.2.5
Define ad hoc binary relation ... 117
4.1.2.6
Define instance attributes ... 117
4.1.2.7
Create class attributes table ... 118
4.1.2.8
Define constants ... 118
4.1.2.9
Define formal axiom ... 119
4.1.2.10
Define instances ... 119
4.1.3
Conceptual Model of OIS Ontology ... 119
4.2
B
UILDING
C
OMPUTATIONAL
M
ODEL
­
F
ORMALIZATION
: ... 121
4.2.1
Actors ... 123
4.2.1.1
Person ... 124
4.2.1.2
Institution ... 125
4.2.2
Domains ... 126
4.2.3
Kinds ... 126
4.2.4
Practice ... 127
4.2.4.1
Information Service ... 128
4.2.5
Studies ... 129

XIV
4.2.5.1
Information economics studies ... 129
4.2.6
Mediator ... 130
4.2.7
Methods ... 131
4.2.8
Resources ... 132
4.2.9
Tools ... 133
4.2.10
Philosophy and theories ... 133
4.2.11
Legislation ... 133
4.2.12
Societal ... 134
4.2.13
Time ... 134
4.2.14
Space ... 134
4.2.15
OIS Components ... 136
4.2.15.1
Classes ... 136
4.2.15.2
Axioms ... 142
4.2.15.3
Properties ... 143
4.2.15.4
Individuals ... 148
4.2.16
Usage Class Tab ... 149
4.3
O
NTOCOP
-
A SYSTEM OF VISUALISATION OF
IS
KNOWLEDGE
... 151
4.3.1
System Requirements ... 151
4.3.2
System Architecture ... 151
4.3.3
System implementing ... 152
4.3.3.1
Technical features... 152
4.3.3.2
Aesthetic Features ... 153
4.3.4
System developments ... 153
4.3.5
Description and potentials of Ontocop components ... 157
4.3.6
Summary ... 160

XV
PART 4: RESULTS & DISCUSSION ... 161
5
CHAPTER 5: RESULTS AND DISCUSSION ... 162
5.1
R
ESULTS
... 162
5.1.1
Evaluation OIS ontology ... 162
5.1.1.1
Ontology validation ... 162
5.1.1.2
Ontology verification ... 166
5.1.1.3
Use case scenario of evaluation ... 166
5.1.1.4
Results of Evaluation ... 169
5.1.2
Results of Ontocop System ... 175
5.2
D
ISCUSSION AND
A
NALYSIS
... 177
5.3
R
EVISED
OIS
MODEL
... 184
PART 4: CONCLUSION & FUTURE WORK ... 187
6
CHAPTER 6: CONCLUSION & FUTURE WORK ... 188
6.1
C
ONTRIBUTIONS
... 188
6.2
A
CHIEVEMENTS
... 190
6.3
F
UTURE WORK
... 191
BIBLIOGRAPHY ... 194
APPENDICES... 205
A.
E
VALUATION
R
EPORT
... 205
B.
T
AXONOMY OF
IS ... 205
C.
G
LOSSARY
... 205
D.
I
NVITATION
L
ETTER
O
NTOCOP
... 205
E.
I
NFORMATION ABOUT PARTICIPENTS PROCESS
... 205
F.
L
IST OF
O
NTOCOP
'
S PARTICIPENTS
... 205
G.
G
ETTING INITIATION OF PARTICIPENTS PROCESS
... 205

XVI
H.
E
XAMPLES OF A COLLECTED
D
ATA
... 205
I.
L
ETTER SENT TO PARTICIPANTS
... 205
J.
R
ESPONSE EMAILS FROM PARTICIPENTS
... 205
K.
E
VALUATION
T
AXONOMY
... 205
L.
P
ART OF
OIS
ONTOLOGY IN
OWL
FORMAT
... 205
M.
L
ESSONS LEARNED
... 205

XVII
List of Figures
F
IGURE
1-1
T
HESIS ORGANIZATION
... 11
F
IGURE
2-1
THE MEANING TRIANGLE
... 19
F
IGURE
2-2
R
ELATIONS BETWEEN TERMS IN
T
HESAURUS
... 21
F
IGURE
2-3
S
IMPLE
T
AXONOMY
... 22
F
IGURE
2-4
SPECTRUM OF ONTOLOGY
. ... 23
F
IGURE
2-5
I
LLUSTRATION OF MIDDLE
-
OUT APPROACH
... 25
F
IGURE
2-6
G
UARINO
'
S PROPOSAL FOR ONTOLOGY
M
ODULARIZATION
... 28
F
IGURE
2-7
AN
E
XAMPLE OF ONTOLOGY ROLE
... 34
F
IGURE
2-8
CONCEPTUAL MODELLING
... 44
F
IGURE
2-9
S
EMANTIC WEB LANGUAGES
... 51
F
IGURE
2-10
SEMANTIC NET IN
RDF,
RDF
S
... 53
F
IGURE
2-11
I
NFORMATION
S
CIENCE RELATIONS WITH OTHER SCIENCES
... 66
F
IGURE
2-12
KNOWLEDGE MAP OF
I
NFORMATION
S
CIENCE
... 67
F
IGURE
2-13D
IFFERENCES BETWEEN
D
EWEY
&
R
ANGANTHAN CLASSIFICATION
... 70
F
IGURE
2-14
CLASSIFICATION OF KNOWLEDGE IN ONTOLOGICAL DIAGRAM
... 77
F
IGURE
3-1
THE MAIN COMPONENTS OF
OIS
ONTOLOGY
... 91
F
IGURE
3-2
TAXONOMY OF
L
IBRARY
S
CIENCE MODULE
... 93
F
IGURE
3-3
D
OMAIN ONTOLOGY OF
OIS
DEVELOPING PROCESS
... 99
F
IGURE
4-1
THE MAIN COMPONENT OF
IS
DOMAIN
... 103
F
IGURE
4-2
SCREENSHOT OF
IS
G
AZETTEER
... 105
F
IGURE
4-3
ANNOTATIONS OF
IS
TERMS
... 107

XVIII
F
IGURE
4-4
ANNOTATION OF
IS
CONCEPTS
... 108
F
IGURE
4-5
RESULT ACCURACY
... 109
F
IGURE
4-6
C
ONCEPTUALISATION ACTIVITIES
... 109
F
IGURE
4-7
SHOWS
T
OP
-D
OWN METHOD
... 111
F
IGURE
4-8
CONCEPT
T
AXONOMY OF
OIS
ONTOLOGY
... 112
F
IGURE
4-9
B
OTTOM
­
UP METHODS
... 113
F
IGURE
4-10
FRAGMENT OF
OIS
T
AXONOMY
... 115
F
IGURE
4-11
AD HOC BINARY RELATIONS
... 116
F
IGURE
4-12
PART OF CONCEPTUAL MODEL OF
OIS
ONTOLOGY
... 120
F
IGURE
4-13
U
PPER
-
LEVEL OF
OIS
ONTOLOGY
... 122
F
IGURE
4-14
M
AIN
A
CTORS CLASS
... 124
F
IGURE
4-15
P
ERSON CLASS
... 124
F
IGURE
4-16
I
NSTITUTION
C
LASS
... 125
F
IGURE
4-17
D
OMAINS
C
LASS
... 126
F
IGURE
4-18
K
INDS CLASS
... 127
F
IGURE
4-19
P
RACTICE CONCEPTS
... 128
F
IGURE
4-20
I
NFORMATION SERVICE CLASS
... 129
F
IGURE
4-21
I
NFORMATION
E
CONOMICS STUDIES CLASS
... 130
F
IGURE
4-22
M
EDIATOR CLASS
... 131
F
IGURE
4-23
M
ETHODS CLASS
... 132
F
IGURE
4-24
R
ESOURCES CLASS
... 132
F
IGURE
4-25
PHILOSOPHY AND THEORIES CLASS
... 133

XIX
F
IGURE
4-26
VISUALIZING
OIS
BY
OWLV
IZ
... 135
F
IGURE
4-27
METHODS OF DEFINING CLASS IN
OWL ... 137
F
IGURE
4-28
DEFINED CLASS IN OWL
... 141
F
IGURE
4-29
AN EXAMPLE OF DISJOINT CLASS
... 142
F
IGURE
4-30
OBJECT PROPERTIES
... 143
F
IGURE
4-31
I
NVERS
O
F RELATION
... 145
F
IGURE
4-32
I
NDIVIDUALS OF
OIS
ONTOLOGY
... 148
F
IGURE
4-33
PROPERTIES ASSERTIONS OF
OIS
ONTOLOGY
... 149
F
IGURE
4-34
USAGE CLASS TAP IN PROTÉGÉ
... 150
F
IGURE
4-35
WEBSITE LAYOUT
... 152
F
IGURE
4-36
O
NTOCOP
F
ORUM
... 154
F
IGURE
4-37
O
NTOCOP CHATTING PAGE
... 154
F
IGURE
4-38
:
O
NTOCOP CHATTING PAGE
... 155
F
IGURE
4-39
O
NTOCOP
M
EMBERS LIST
... 155
F
IGURE
4-40
O
NTOLOGY
P
AGE
... 156
F
IGURE
4-41
F
EEDBACK
P
AGE
... 156
F
IGURE
4-42
FQA
P
AGE
... 157
F
IGURE
4-43
C
ONTACT
P
AGE
... 157
F
IGURE
5-1
CIRCULAR CLASSES
... 164
F
IGURE
5-2
INFERRED CLASS HIERARCHY
... 165
F
IGURE
5-3
PART OF
OIS
ONTOLOGY VERIFICATION RESULTS
... 166
F
IGURE
5-4
E
VALUATION OF
IS
TAXONOMY
... 167

XX
F
IGURE
5-5
SNAPSHOT OF
OIS
ONTOLOGY ON
W
EB
P
ROTÉGÉ
... 168
F
IGURE
5-6
OIS
DOCUMENTATION
... 168
F
IGURE
5-7
ONTOLOGY CONSISTENCY
... 170
F
IGURE
5-8
C
ONSISTENCY OF IS
-
A AND PART
-
OF
­
RELATIONSHIPS
... 171
F
IGURE
5-9
COMPLETENESS OF ONTOLOGY
... 171
F
IGURE
5-10
C
LARITY OF
OIS
ONTOLOGY
... 172
F
IGURE
5-11
ONTOLOGY GENERALITY
... 172
F
IGURE
5-12
SEMANTIC DATA RICHNESS OF THE ONTOLOGY
... 173
F
IGURE
5-13
T
HE
G
ENERAL ASSESSMENT ON
OIS
ONTOLOGY
... 175
F
IGURE
5-14
PARTICIPANTS OF
O
NTOCOP
... 175
F
IGURE
5-15
VISITORS OF
O
NTOCOP
... 176
F
IGURE
5-16
SATISFACTION LEVELS WITH THE
OIS
ONTOLOGY
... 181
F
IGURE
5-17
EVALUATION CRITERIA AT LEVEL
3 ... 182
F
IGURE
5-18
S
EARCHING ON
W
EB
P
ROTÉGÉ
... 184
F
IGURE
5-19
ONTOLOGY MATRICES
... 185
F
IGURE
6-1
A
RCHITECTURE OF SYSTEM DESIGN APPROACH
... 189
F
IGURE
6-2
I
NTERFACE OF
OIS
ONTOLOGY SEARCHING
... 192
F
IGURE
6-3
R
ELATIONSHIPS BETWEEN ONTOLOGIES
... 193

XXI
List of Tables
T
ABLE
2-1
SIMILARITY BETWEEN ONTOLOGY AND CATEGORY THEORY
... 17
T
ABLE
2-2
D
IFFERENCES BETWEEN TAXONOMY AND ONTOLOGY
... 23
T
ABLE
2-3
ONTOLOGY CATEGORIES
... 30
T
ABLE
2-4
D
OMAIN ONTOLOGIES
... 33
T
ABLE
2-5
A
PPROACHES OF ONTOLOGY EVALUATION
... 36
T
ABLE
2-6
AN OVERVIEW OF LEVELS OF ONTOLOGY EVALUATION
... 39
T
ABLE
2-7
IMPLEMENTATION OF
M
ETHONTOLOGY
... 45
T
ABLE
2-8
COMPARISON BETWEEN METHODOLOGIES
... 48
T
ABLE
2-9
M
ETHODOLOGY
S
TANDARDS
... 49
T
ABLE
2-10
C
OMPARISON BETWEEN SEMANTIC LANGUAGES
... 54
T
ABLE
2-11
OWL
CONSTRUCTORS
... 56
T
ABLE
2-12
OWL
AXIOMS INTERPRETATION AND FACT
... 57
T
ABLE
2-13
COMPARISON BETWEEN ONTOLOGY LANGUAGES
... 58
T
ABLE
2-14
G
ROUPS OF
O
NTOLOGIES
T
OOLS
... 58
T
ABLE
2-15
C
OMPARISON BETWEEN ONTOLOGY TOOLS
... 62
T
ABLE
2-16
COMPARISON BETWEEN COMMUNITIES OF PRACTICE
... 87
T
ABLE
4-1
THE SCOPE OF
IS
DOMAIN
... 103
T
ABLE
4-2
PART OF THE GLOSSARY OF TERMS OF
OIS
ONTOLOGY
... 110
T
ABLE
4-3
CONCEPTS CLUSTERING
... 114
T
ABLE
4-4
CONCEPTS DICTIONARY
... 117
T
ABLE
4-5
PART OF THE AD HOC BINARY RELATION OF
OIS
ONTOLOGY
... 117

XXII
T
ABLE
4-6
SHOWS PART OF INSTANCE ATTRIBUTES OF
OIS
ONTOLOGY
... 118
T
ABLE
4-7
A
SECTION OF THE INSTANCE ATTRIBUTES TABLE OF
OIS
ONTOLOGY
... 118
T
ABLE
4-8
A SECTION OF CONSTANTS TABLE OF
OIS
ONTOLOGY
... 118
T
ABLE
4-9
THE INSTANCE TABLE OF THE
OIS
ONTOLOGY
... 119
T
ABLE
4-10
T
YPES OF RELATIONS BETWEEN TERMS
... 144
T
ABLE
5-1
INCONSISTENCE CLASSES
... 163
T
ABLE
5-2
T
HE QUESTIONS IN
OIS
ONTOLOGY SURVEY
... 169
T
ABLE
5-3
LEVEL
3
OF SATISFACTION ON ONTOLOGY BASED ON SPECIFIC CRITERIA
... 182

XXIII
Acronyms
AI
...Artificial Intelligence
ANNE
...A Nearly New Information extraction System
CLIPS
...C Language Integrated Production System
COC
...Community of Commitment
CoI
...Community of Interest
Cops
...Communities of practice
DARPA
...Defence Advanced Research Projects Agency
GATE
...General Architecture for Text Engineering
IE
...Information Extracting
IS
...Information Science
JAPE
...Java Annotation Patterns Engine
KE
...Knowledge Engineering
KR
...Knowledge Representation
LOOM
...Language for Object Oriented Methods
Nop
...Network of practice
OCML
...Operation Conceptual modelling Language
OIS
...Ontology of Information Science
Ontocop
...Community of Information Science website
OWL
...Web Ontology Language
RDF
...Resource Description Framework
SGML
...Standard Generalized Mark-up Language
VCops
...Virtual Community of practice

XXIV
VCs
...Virtual community
W3C
...World Wide Web Consortium
XML
...Extensible Mark-up Language

1
Part 1: Fundamental Issues

2
1 Chapter 1: Introduction
Recently, the development of domain ontologies has become increasingly important for
knowledge level interoperation and information integration. They provide functional
features for AI and knowledge representation. Domain Ontology is a central foundation
of growth for the semantic web that provides a general knowledge for correspondence
and communication among heterogeneous systems. Particularly with a rise of ontology in
the artificial intelligence (AI) domain, it can be seen as an almost inevitable development
in computer science and AI in general.
Ontologies are useful for different applications to be able to share information between
heterogeneous data resources. They are also essential for enabling knowledge-level
interoperation of agents, when these agents are interacting to share a common
interpretation of the vocabulary. Moreover, it is useful for human understanding and
interaction to reach a consensus amongst a professional community.
Although there are a range of domain ontologies on the semantic web such as Gene
Ontology (GeneOntology, 2009), Biological science ontology (Sabou 2005), CIDOC-CRM
ontology of culture heritage documentation, FRBR in Bibliographic and NCI cancer
ontology (Golbeck et al., 2008), there still exists a lack of domain ontologies, which has
led to the loss of knowledge in specific domains. This is a significant problem for scholars
and researchers who need to be able to access information within their interest area.
Ontology provides a vocabulary for metadata description with machine understandable
terminology. Ontology provides a format for explaining and understanding terminology
and the knowledge contained in a software system. By using shared concepts and terms
in accordance with a specific approach, a lot of information remains in people's heads. It
is discussed in 2.3.
However, information science (IS) is a fast paced discipline and communication
technology is rapidly increasing, so it is imperative to take advantage of this
development. IS is a multidisciplinary field and it has gained the fundamental root of its
theory from different related fields. The analysis includes the three branches of the field,
which are; Library Science, Archival Science and Computer Science. Meanwhile it
overlaps with other sciences, as stated in Section 2.2, e.g., communication, cognitive
science, philosophical science, management, social science and marketing. More
precisely, the relationships between information and marketing can be subdivided into
marketing information, marketing information services, marketing of library services.

3
These kinds of relationships need logical ontology to clarify their relations and the
science boundaries, amongst others. Therefore, Information Science still needs identity.
However, there is a lack of IS ontology representing the unified model that combines all
concepts and their relationships. Moreover, IS as any domains which use the natural
language. It contains a lot of jargon which needs to be in a formal language for
programming or logic. Alternatively, integration of the computer with the internet has
led to the emergence of new concepts in the field of IS such as , Electronic Library,
Virtual Library, Library Without Walls, Digital Library and Information Management, as
well as Nerve Centres. Even the information concept itself has strong and complex
relations with other concepts, for example some people have defined it as fact, energy,
data, and symbols. Also, it can be composed with other words such as; information age,
information revaluation, information crisis, information explosion. However, there are
400 definitions for information in the literature (Yuexiao, 1988). It is hard to differentiate
between these concepts. Even within the same field, there is still confusion over defining
information - everyone defines it based on his background, for example librarians know
it in term of facts, and data can be in containers such as journals, books and documents.
The computer scientists conceive it as small units such as bits and bytes.
Consequently, modelling the IS domain necessarily assumes the need to represent the
correct picture of the whole domain, and any changes in the domain will have to be
added to keep the model up to date (Mommers, 2010, Yuexiao, 1988).
Our consideration is that in developing an ontology of Information science OIS to define
its boundaries, and avoid ambiguous concepts.
Therefore, there is a lack of unified model of domain knowledge, because of the
inconsistency in structure of domain which led to difficulty of using and sharing data in
syntax and semantic level.
1.1.
Problem Identification
Information Science is seeking its identity and it is one of the many domains which use
natural language including much jargon. Also, integration of the computer with the
internet has led to emerging concepts in the field of IS such as , Electronic Library,
Virtual Library, Library without walls , digital Library It is hard to differentiate between
them.
Furthermore, its structure led to lack of a unified model of domain knowledge. This led to
lack of a unified model of domain knowledge, and difficulty of using and sharing data at

4
syntax and semantic levels. The OIS ontology provides a standard terminology and
shared representation of domain concepts.
Therefore, the ontology of information science is missing in ontological engineering area.
Our consideration is that developing ontology of Information science to define its
boundaries, and to avoid the concepts ambiguous.
The research problem of the study was defined as the following:
Q. How an ontology of Information Science (OIS) model can be
developed to visualise the IS domain, and how the model could capture
and represent this knowledge?
To achieve the primary objective, the researcher asks questions to be answered through
this study such as:
-
What domain knowledge does the ontology represent?
-
What is the level of knowledge that the ontology will represent?
-
Which knowledge representation techniques and languages should be used?
-
What are the relations that will be used to structure the knowledge, and which
structure for the ontology will it have e.g. tree, graph, and its main components
of ontology (e.g., classes, instances, relations, rules)?
-
What is the value of tools such virtual community of practice ontocop? Could
they be valuable in supporting the developing process?
-
Does the developing process of the ontology follow designing criteria?
-
Is the ontology evaluated based on specific criteria?
1.2.
Aims and Objectives
The aim of this research is to develop a generic model of ontology that visualize domain
knowledge of IS that serves as a foundation of knowledge modelling for applications and
aggregation with other ontologies.
The visualisation stage provides an extensible and commonly understood semantic
framework by describing the terminology of the domain. Achieving this aim in the
current study will fulfil the following Objectives:
-
Building a conceptual model for establishing a better analysis framework to
understand, classify and compare various classes of Information Science.

5
-
providing a framework to make it possible to share a common understanding of
Information Science by:
o Identifying the key objects of IS domain and relationships.
o Providing a specification of information requirement for both developers
and end users, to be used in different applications.
1.3.
Methodology and Implementation
The aim of this part is to investigate whether the results found in the literature study
could be applied in practice by focusing on ontologies in a specific area. For this purpose,
the virtual community of practice (Ontocop) was designed to visualise the area of
Information Science (IS). Also, to involve other people as member of VCops by using
some process of negotiation, to give us feedback on the ontology it is been developed.
Additionally, they will help the researcher to assist and evaluate the ontology. There are
many different methods for asking for feedback and analysis what the results are.
The literature review will be used in this research to address the research problem as
identified by Saunders, et al (2000). It will be include the key of academic theories
through the chosen area, and revealing that knowledge of your chosen area is new.
Beside explain how the research relates to previous published research, to justify
arguments by referencing prior works. Furthermore, enabling readers to find the original
work you cite through apparent reference.
Regarding building the ontology, a methodology for building ontologies decides the main
development stage and proposes guidelines for each stage dependent on use of the
ontology. Many methodologies have been proposed since the 1990s to build ontologies.
Each one has a different approach, such as Methontology, and SENSUS. Gòmez-Pérez et
al. (2004) have made comparisons between these methods, and have pointed out that
these methods have common development stages most of them have conceptualisation,
requirements analysis, formalisation, implementation, maintaining and evaluating.
Hence, there seems to be no general agreement on methodology to building and design
ontology, due to the fact that it depends on its application and purpose of using it Noy
and McGuinness (2009). To build a new ontology from scratch, or reuse another
ontology, it should be built according to present needs and the purpose from it (Pinto
and Martins 2001).
In this study, a new approach is proposed for designing a system to build ontology
through sharing and reusing knowledge between members of communities of practice of
Information Science (IS). The first step is building the ontology through the (VCops).

6
The second step is building ontology of Information Science (OIS). In this sense our
approach to visualise the knowledge of IS domain, will be as depicted in Chapter 3.
1.4.
Contributions
In this research the main contribution presented through this thesis is:
-
Creating ontology of Information Science OIS model to unify IS knowledge. The
OIS ontology is a general model for the domain, enabling the integration of a
large amount of information resources. It designed to be flexible, reusable for
other implementations, and compatible in knowledge base systems rather than
imposing a specific solution.
-
The model has fundamental roots in a framework based on analysis of the
knowledge of IS domain; our framework is to identify the domain boundaries and
relationships among them by providing IS taxonomy. Although there are many
classification systems in the world none of them represent this in a formal way. In
this study OIS taxonomy will be represented in OWL formal presentation; the
taxonomy approach is described in Section 3.1.1.
The model has defined 706 concepts which will be widely used in Information
Science applications. It provides the standard definitions for domain terms used
in annotation databases for the domain terms, and avoids the consistency
problems caused by various ontologies which will have the potential of
development by different groups and institutions in the IS domain area.
-
Design VCops (Ontocop) to support and assess the development process as
specific virtual community of IS. The Ontocop consists of a number of experts in
the subject area around the world. Their feedback and assessment improve the
ontology development during the creating process.
The structured ontology was developed as a specific model of IS domain by following
Methontology based on the IEEE standard (1996, 2006) for development software life
cycle process. It mainly consists of the four main stages described in Section 3.3. The
methodology and tools of design ontology was determined based on the experiments of
Uschold and Grüninger M.( 1996), Noy and McGuinness, (2001).
The designing evaluation tool is presented in Section 4.1. The research tools adopted
were;
1. Design a virtual community of practice (ontocop) evaluation ontology model.

7
2. The study used information extraction (IE) techniques to annotate the key
entities of IS using JAPE grammar and General Architecture for Text Engineering
(GATE) for data annotation; more details can be found in Section 4.1.1.2.
The principle resources that have been used are domain experts through Ontocop, who
were consulted to assess the ontology based on their experience and knowledge.
This research attempts to improve understanding of the distinctions among information
science as a whole. Therefore, it is seeking to describe the constituents of the IS field,
and ideally to put these into set theoretical foundations in Section 3.1.
The research does not provide any a priori assumptions of using precise details about
the IS domain, insofar as it is a generic model intended to provide a control vocabulary
that can be applied for IS applications. It is important to note that the ontology model
does not cover the range of individuals and extending relations. Nevertheless, it defines
the concepts that serve as the foundation of IS, such as Actors, Methods, Domains, etc,
which need to be extended in future use with corresponding ontologies.
OIS ontology is structured as a combination of domain and an upper ontology. The upper
ontology contains a foundation of the ontology. It offers very general entities with
subclasses, attributes, objects that
give potential sources of integration with other
ontologies. The IS domain has a strong relation with others.
The reason behind that, however, is that the domain ontology presents specific concepts
of the domain in eclectic ways, which are often incompatible and incomplete. These kind
of ontologies need to be merged and shared with other ontologies into more general
representation. Also, it should be well-matched to the equivalent semantic area with
corresponding ontology. Particularly in the IS domain, this consists of a complex
combination. By using a common foundation, ontology provides basic elements for
emerging domains ontologies automatically. The ontology model is a comprehensive
scope covering three branches that are closely related to the domain; library science,
archival science, and computer science.
The purpose of the OIS model is not to serve a broad spectrum of librarians, academic
staff, publishers, information service providers only insofar it takes into account a variety
of applications. Entities, relationships and attributes are the basic components of the
model; these elements were derived from logical analysis of IS data.
Furthermore, the research describes the strategy and method developed to build the
domain ontology of IS. It believes that this research offers significant advantages to

8
modelling domain knowledge, in term of the contents of developing the IS ontology. This
study created domain ontology and it is not considered task and application ontology.
The main purpose of the OIS ontology is to provide a unified model of domain knowledge
that supports knowledge sharing and the exchange of data among databases.
1.5.
Motivation of study
Ontology is not just identifying classes as entities and their relations and concept
hierarchy but also specifying them by using specific ontology representation languages.
OIS ontology seeks to provide a formal model of Information Science domain that is
formulated in description logic. OIS ontology aims to represent domain knowledge to use
independently of any application.
The motivation will be therefore at these possible levels:
-
Ontologies represent knowledge about the real world. Nowadays, with growing
attention to ontology, IS needs ontology. The problematic situation is
identification of IS itself, especially the overlap between it and library science,
computer science and archives science. On the other hand, there are many
attempts to change the identity of the science to Knowledge science rather
IS(Zins, 2007a). From this perspective we need a serious attempt to challenge
the identity of IS through identification of its boundaries and relations with other
fields, through this research, in Section 2.2.
-
Information Science just as any other scientific field requires a framework for
organising its knowledge, especially with the fast speed of development
disciplines. The terms data information and knowledge still have definition issues,
although there have been many attempts to define and distinguish between
them, precise definition is still problematic (Zins, 2007a, Wiederhold, 1986,
Bubenko and Orci, 1989).
-
Providing a consensual knowledge model of the IS field to be used by application
ontologies. Hence, developing ontology enables the application to manage
complex and disparate information. Also, changing the semantic web structure
from surface composition to be captured in the application logic.
-
Using a virtual community of practice as a way of sharing knowledge. Although
there has been extensive discussion about the use of communities of practice in
this way, no formal academic research has been identified relating specifically to
the context of evaluating ontology via VCops in the Information Science domain.

9
1.6.
Thesis organization
This research is structured into 5 parts and 6 Chapters. Each part is preceded with a
brief introductory section to explain how the work presented in the Chapter fits in the
overall structure of the research.
Part 1: presents the fundamental issue of the research. Chapter 1 provides the
identification of the problem, and the aims, objectives and motivation of the study,
research methods, and research organisation. Furthermore, to solve the problem
identified we create ontology in Chapter 4.
The second Chapter 2 presents a survey of the previous studies to provide contextual
information on the main components of the research; Section 1, which is about ontology
for semantic web overview, presents the origin of the ontology, and introduces the
formal definition of an ontology that supports the communication between human and
machine. Additionally, it introduces types of ontologies and provides techniques to
represent ontologies based on web standards languages e.g., XML, RDF. Furthermore, it
presents a comprehensive framework of ontology layered for the semantic web. Section
2 is about Information Science as domain of the ontology, as well as Section 3 which is
about knowledge management and virtual communities of practice. This part explores
related work to provide the background to the research.
Part 2: presents methodology of creating ontology of Information Science OIS in
Chapters 3 in two sections. The Section 1 provides the theoretical model of the current
study. The Section 2 presents the methodology that has followed in the research to
design OIS ontology.
Part 3: implementation of OIS ontology model Chapter 4 in two sections, which
provides the functionality of the implemented tool environment for ontology engineering
called Protégé. It provides a numbers of screenshots and examples of the running
system. Section 1 presents the model design and Section 2 presents the ontocop system
design.
Part 4: Results and Discussion, in Chapter 5, has two sections. Section 1 provides our
approaches to evaluating OIS ontology. These approaches are based on a number of
ontology quality criteria, to consider the question of how this Information Science
ontology will be used and whether it will be useful, and if the answer is yes, which
context or application ontology will use it, as identified in Chapter 4.

10
Part 5: consists of one Chapter 6 of conclusion and future work, to draw together the
contributions this research offers, and a direction to future work. Figure 1 gives a
graphical overview of how to construct the research.
Appendices are at the end of the thesis. The evaluation report is found in Appendix A.
Appendix B includes Taxonomy of IS. Appendix C includes the Glossary of IS terms. The
ontocop collection is included in Appendix D, which also contains the invitation letter to
invite participants to ontocop, and E contains information about participation process.
The members list is found in Appendix F; Appendix G is about getting initiation of
participant's process ­ it explains how members can start using the ontocop. Appendix H
contains examples of the database of participants; Appendix I contains the letter of
setting at ease starting of participants, and Appendix J contains the response emails of
agreement to the participation. Appendix K consists of the feedback on evaluation
taxonomy. Appendix L is a part of OIS ontology OWL file. Finally, some lessons learnt
during the study can be found in Appendix M.

11
Figure
1-1 Thesis organization
Part1
Fundamental Issues
Part2
Methodology of creating
ontology of Information
Chapter 1 Introduction
Chapter 3 Method Employed
Section 1 Theoretical
approaches
Section1 Ontology
for semantic web
Section 2
Ontocop system design
Chapter 4 Section 1
Modelling of OIS design
Section 2
Methods of Model Design
Section3
Knowledge
management and virtual
Chapter 2 Research
Background
Part3
Implementation
Part5
conclusion & future work
Chapter 6
Conclusion & future work
Section2 Information
Science domain
Part4
Results & Discussion
Chapter5 Section 1 results of
Ontology evaluation
Section 2 Discussion

12
2 Chapter 2: Research Background
The literature review gives the background to the research process, which consisted of
three main aspects to find out the theoretical background essential to this project. These
aspects were: ontological engineering, Information Science, and Communities of Practice
within knowledge management. The following sections provide an overview of key
literature relevant to this project.
Firstly, however, the background starts with some basic definitions to establish what is
meant by ontology and what the significance of creating ontology is. The survey will
come back to the three key aspects of this study and review literature on these; firstly,
ontology.
2.1
Ontology Overview
Ontology plays an important role to use as a source of shared defined terms ­ for
instance metadata ­ which can be used in a specific domain (Gaoyun et al., 2010). The
concept of ontology became popular in the 1990s. Ontology's meaning can change
according to the context of where it is used ­ for instance in philosophy, computers,
linguistics, mathematics or social science. It is defined differently in work relating to
computer science. Barry Smith (2003) said that ontology is a science of the existence of
beings, and as such it has a relationship with computer and information science as a
field.
Interest in the area of ontology in computer science has grown in recent years (Amira et
al., 2007, Bhatt et al., 2009). In the early 1990s, ontology definitions as a term within
computer science emerged. Computer science defined ontology based on knowledge
systems (KMS) as a classification of knowledge (Guarino, 1997).
Ontology has a long history of development which predates computer science. This
section will begin by reviewing the historical background of ontology, and the
philosophical perspective will be introduced. Then, moving forward to defining ontology
based on comparing the original use with its current use in computer science will be
combined, which will lead to a formal definition of an ontology that will be the basis for
this research. Then, the thesis will move on to describe the development of ontology and
share an explanation of the benefits of developing ontologies. It summarising
approaches to modelling ontology with some examples of ontologies. Finally, we
summarise some methodology, and explain the tools such as Protégé and the languages
used for representing ontologies.

13
2.1.1
Historical and philosophical perspective of the ontology
To understand the ontological foundation for the ontology of Information Science it
required reviewing diverse approaches to the notion of this concept. This section reviews
some of the literature that is relevant to philosophical ontology. We explore some views
from logicians that have influenced this project.
The ontology concept came from a branch of philosophy. Philosophers used ontology as
a synonym of metaphysics - that means anything comes after the physical (Smith,
2003). Consequently, they defined it as a theory related to the study of relationships
between beings (Webster's, 2010).More accurately, ontology is the study of things
categories that may exist or already do exist in some domains (Sowa, 2000).
Back to the history from a philosophical perspective, Aristotle (384-322BC) invented
ontology as a study of the ways that the universe is organised into categories. The
category is the highest level of universal obtained from those domains; all other
universals reorganised their hierarchies that need the top levels of categories, such as
City, Man, and Organism. In (1200-1600) medieval scholars developed a common
control vocabulary for talking about these universals in terms of sorts of reality.
Descartes only initiated a movement of epistemology as a centre of philosophy rather
than ontology or metaphysics until around (1960-61) by differentiating between mental
and physical subspecies which had not been a problem for Aristotle. Brentano (1838-
1917) denied the differences between philosophy and science; he said they are one and
the same. Husserl (1859-1938) influenced by Brentano, invented formal ontology as a
discipline distinct from formal logic. He showed how philosophy and science had become
detached from the real life world or ordinary experience (Calero et al., 2006).
Philosophical ontology is a way of describing reality by providing a comprehensive
classification of entities. That means organising all kinds of relations by classes or
entities collectively (Merrill, 2011).
In general, methods of philosophical ontology are derived from philosophical methods.
These methods include theory development, and testing and modifying them.
Furthermore, these methods were similar to Aristotle's view.
Many philosophers had made distinctions between logic, computation models and
ontology. Robert Poli (2003) has discriminated further between Husserlian formal
ontology, descriptive and formalized ontologies. This distinction appeared from
discussion of the main role of logic in these formalisms of ontology. Husserl's logical view
had asserted that logic is an essential part of formal ontology (Poli, 2003). The group of

14
AI has followed this theory where the formal ontology contained concepts, logical
axioms, theorems and mereology. However, according to Tim Berners-Lee's semantic
web tower, logic is the top layer above ontology vocabulary (BERNER-LEE, 2001). More
interestingly the technical and knowledge representation aspects have been using a
robust concept of Web Ontology Language (OWL) as W3C recommendations are based
on the description logic.
Recently, ontology has become associated with AI and information systems. AI logicists
have focused attention on the knowledge-based craft. In 1980 McCarthy recognized the
overlap between philosophical ontology and building logical theories of AI systems.
McCarthy (1980) confirmed that developers of logic based on intelligent systems need to
accumulate everything that exists to build the ontology.
Nirenburg and Raskin (2001) emphasize that ontological semantics is a theory of
meaning in a Natural Language Process (NLP) that supports many applications such as
information extracting and machine translation. Crucially, however, a good ontology
requires choosing concepts that have to be covered and reasonably consistent. The
ontology designers decide how to arrange and organise the concepts to be included
(Nirenburg and Raskin, 2001, Nirenburg and Raskin, 2004).
In the interim, a similar view of overlap with philosophical ontology was proposed by
Joan Sowa; ontology is to be considered as catalogue for possible global use that puts
everything together and defines how it works (Sowa, 1984).
The AI community prefers to use the concept of ontology in knowledge engineering
without much overlapping with the field of philosophical ontology. They work under the
title of "ontology" that is related to logical semantics and logical theory.
Alexander et al., (1986) initially used the concept in the AI sense. This concept has been
grown considerably in different fields of Database Management Systems (DBMS),
knowledge engineering, domain modelling and conceptual modelling.
2.1.1.1 Definition of ontology
Since the AI community discovered the power and knowledge within their systems,
ontologies can refer to an engineering artefact to present a formal specification
developed with AI, or an informal specification for human users. The AI community
defined ontology as:

15
"Ontology is a theory of what entities can exist in the mind of a knowledgeable agent".
(Wielinga and Schreiber, 1993)
In 1993 Tom Gruber coined the concept Ontology in a sub-field of computer science.
Gruber gave us the most widely-shared definition of ontology as a conceptual model:
"An ontology is an explicit specification of conceptualisation."(Gruber, 1993a)
But his definition has many interpretations, which are that ontology can provide a
specification of conceptualisation of generic notions such as space and time or domain
application. A number of researchers in the computer science community have
attempted to clarify and formalise the ontology definition further such as (Guarino,
1998).
Guarino and Giaretta (1995) highlighted the importance of terminological classification,
to avoid misunderstandings over an ontology as a conceptual framework at knowledge
level and an ontology as an artefact at symbol level, used for a specific purpose. The
concept was further developed in 1999 when Welty and his colleagues described a range
of information artefacts that had been classified as ontology. (Welty et al., 1999)
Meadche (2002) defined ontology formally as containing classes, relations and axioms,
whilst also allowing for lexical entities referring to multiple concepts and relationships
(homonym). It also refers to the concepts and relations through several lexical entries
(synonym).
In 1993 Gruber defined ontology as:
"An ontology is a specification of a conceptualization." (Gruber, 1993b)
His definition has been developed to be more accurate for defining ontology which is:
"Formal explicit specification of shared conceptualization"
Ontology makes the term clearer and indicates in which context the term can be used.
The definition consists mainly of:
A formal: ontology should be machine readable and processed by AI systems. We do
not need it to be a communication device between people and people, or even people
and machine. Ontology should be formally defined as a formal language. (Morbach et al.,
2009)
Specification: means written specifications of language syntax to satisfy certain
criteria such as precise, unambiguous, consistent, complete and implementation

16
independent statements (Turner and T.L, 1994). It should offer a communication tool
whereby users can share knowledge in consensual ways.
Shared: ontology represents consensual knowledge that, has been arranged and agreed
on by group of people as result of social networks rather than an individual's view.
Conceptualisation: this is an abstract model of a domain that is driven by user
application, and represents concepts and relationships to be shared and reused.
Conceptualisation is based on objects, concepts and other entities already in existence in
the area of interest.
Based on this, ontology should be formally defined as being processed by a machine.
The ontology is a specific type of information object or artifact. The way the ontology is
constructed refers to classes, relations and their instances, all of which play explicitly
specified roles in the conceptualisation. Otherwise, the backbone of the ontology consists
of specification or generalisation hierarchy of concepts. However, Ontology is not
software, though, so whilst it can be used by programs, it cannot run as a program
A far more interesting question is what information systems could learn from
philosophical ontology. It is a shared belief that there is a similarity inherent in ontology
from philosophical and applied scientific perspectives. Philosophical ontology is
describing the real world as it exists, while computational ontology is describing the
world as it should be (Kabilan, 2007).
2.1.2
Ontology Theoretic
2.1.2.1 Category Theory
A number of thinkers and pioneers as Aristotle, Hartmann and Husserl (Bello, 2010,
Hartmann, 1952), point out that ontology is adopted as a categorical framework that
means it seeks for what is universal (Poli, 2010). Husserl's emphasis on the premise of
the category theory could be reflected in many ways according to different viewpoints
The precise meaning of ontology relies on the theory of category as a grounding in
contemporary mathematics (Lawvere, 1969, Krötzsch et al., 2005, Johnson and
Dampney, 2001, Awodey, 2006, Hu and Weng, 2010).
Similarities in the relationship between category theory and ontological representation
technique are summarised in Table 2-1

17
Table
2-1 similarity between ontology and category theory
similarity
Ontology
Category theory
classification
as Tree
grammars using tree or TAGS
Defining language
present language by defining
term
Mathematical concepts
node
Has node of tree
Has node of tree
relations
Interrelations
Close relations between formal
linguistic presentation of
domain & tree base
representation.
However, categories appear in different ways such as taxonomy (is-a superclass,
subclass), to group the domain in classical taxonomical categories according to Aristotle
perspective. Recently Aristotle framework becomes matter particularly with time.
Theories can help to define formal ontological properties that contribute to characterising
the concepts. Husserl introduced the theory of Mereology as basic for formal ontology,
and it is an alternative of set theory described by Tennant (2007).
2.1.2.2 Mereotopolgy Theory
Mereology is a formal theory concerned with wholes and parts structures (Husserl,
1970), whereas topology is a theory of wholeness that defines the relations connected to
its properties, and how to be represent these components within the system (Varzi,
1996).
The basic metrological system is M= (E, ) in which E is domain entities, and is binary
relations. The E, binary relation is denoted; M can be considered as ground Mereology.
The ground Mereology is the first order partial ordering theory as reflexive,
antisymmetric, transitive relations; some relations can be axiomatised as follows:
(M1) x (x x), (reflexive)
(M2) xy (x y y x
x = y ), (anti-symmetry)
(M3) x y z (x y y z
z ), (transitivity).
More precisely, the general framework Mereology system is defined to the level of
granularity and predicate:
M(D) = (E, wh(x, l), P(x,y))

18
Any domain is introduced M(D), and where/why(x, l) is the level of granularity and
predicate , expressing that x is entity of the level of granularity L.
But with the weakness of this theory it requires more axioms to recomplete the functions
(Varzi, 1996, Herre, 2010) The formal precise theory identifies and describes the
classical first order logic using variables Y, X, Z etc. For the theory to be semantically
and ontologically adequate it is required.
The axioms in Mereotopolgy are designed to serve a formal ontological system. The
primitive relations of parthood or constituency are as follows: if says x is a part of y `x P
y' then y will be consisted with x's being identical to y:
x overlaps y xOy: = z (zPx zPy)
x is discrete from y: xDy: = xOy
x is a point Pt (x): = y (yPx
y =x)
While, Boundaries defined as follows:
×By: = z (zP× z\sty)
If X is tangent y then x T y:= z(zPx zTy)
If X cross y then xXy:= xPy
-xDy (Barry, 1996)
This research is based on (Herre, 2010)'s view about constricting a domain which is:
D=(obj(D), V(D), CP(D).
D is a domain that is determined by set of objects obj(D) connected to it. These objects
rely on a set of views V(D), and a set of classification principles (CP) for objects obj(D).
To make the components highly formal it is necessary to use categories and relations
between them. In this case, the domain should be represent as:
Concepts (D) = Cat(D), Rel(D), Obj(D).
It is based on (Gurbe, 1993)'s approach of specification of conceptualization. The domain
components are supported by relationships Rel(D), classification principle- taxonomy
CP(D), additionally the concepts of the domain will be determined by adding axioms,
these axioms are presented by interrelations between categories and its properties.

19
2.1.3
Referencing and meaning in the ontology
Human communication theory is expressed in a general communication context using
the triangle of meaning. As depicted by Ogden et.al (1949) this contains three
relationships between words, thoughts and things. This describes the real world
interaction between thoughts (concepts), words (terms) and things (objects), as
depicted in Figure 2-1.
The diagram shows the relationship between objects and concepts, and an indirect
relationship between terms and objects, meaning there is no matching between words
and things. In natural languages such as Spanish or English, each concept has a
meaning. To explain further, a concept often carries more than one meaning, based on
the knowledge background and historical structure in an individual's mind; for example,
if someone talks about "AAAE5", the person listening to them won't understand them
because there's no matching image in his mind to interpret this or connect it to the real
world. However, when the conversation is about a specific concept, for example
"jaguar", everyone will interpret or imagine it, based on their background knowledge.
One will think it is an expensive car that has an engine, four tyres and needs oil to
move, and so on. The other thinks it is a big cat. In this way, one concept can have
different meanings.
Concepts are a basic part of the proposition. They can express a certain meaning. The
conceptual model helps to abstract models of parts of reality, by describing the key
concepts and their relations.
More interesting than this, however, is what ontology can do in this case as a type of
conceptual modelling method. Ontology attempts to represent the meaning of concepts,
their properties, values and attributes. It provides a clear definition by stimulating a
Figure
2-1 the meaning triangle

20
particular meaning, in this case that a jaguar is a big cat with four legs which lives in
America. Ontology helps to avoid confusion and supports effective communication.
2.1.4
Ontology spectrum
The first task in the ontology of IS is to control the vocabulary being used. The intention
of ontology is to capture and reuse knowledge on a particular subject between software
applications and groups of people (Gómez-Pérez et al., 2003). In reality, the nature of
ontology has many aspects ­ some people consider it a thesaurus, some a data
dictionary, and others a representation of concepts, classifications or taxonomy.
2.1.4.1 Thesaurus
However, the most popular way of controlling vocabulary is the thesaurus, which is a list
of words grouped together, based on their meaning. Librarians in libraries and
information centres use it as a tool to categorise information for the purpose of
information retrieval. A thesaurus is similar to ontology in some aspects:
-
Organizing terminologies in consistent ways.
-
Using hierarchy structure as category and subcategory.
-
Using terms in a particular domain.
-
Providing information as synonym relation.
A thesaurus differs from ontology because a thesaurus provides ambiguity in
relationships and offers alternative words and meanings. (Broader then BT, Narrow then
NT, Related to RT). These relations are offered but they are unclear and aren't formally
defined, unlike ontological relations. The relations should relate to a specific term rather
than a range of terms and should also indicate that this term is a part of another term,
e.g. (A) is subclass of (B) and (D) is a superclass of (A). Furthermore, the relationships
in ontology indicate classes, subclasses, relations and properties, axioms. Ontology
therefore provides far more than relationships. Relating to this Daconta (2003) pointed
out other relations that had parallels with terms in the thesaurus, such as:
-
Equivalence; if term (A) has a synonym then term (B) is equivalent.
-
Homographic; when term (Y) is spelled as (F) but has different meaning.
-
Hierarchical; the term could be narrower than and broader than, e.g.
If (A) is broader than (B); then (A) is superclass of (B).
If (C) is narrower than (D); then (C) is subclass of (D).

Details

Pages
Type of Edition
Erstausgabe
Year
2015
ISBN (PDF)
9783954899487
ISBN (Softcover)
9783954894482
File size
14.6 MB
Language
English
Publication date
2015 (June)
Keywords
Ontology artificial intelligence Knowledge Representation Visualisation Virtual Communities of Practice Information Science Web Ontology Language
Previous

Title: Ontological Engineering approach of developing Ontology of Information Science
book preview page numper 1
book preview page numper 2
book preview page numper 3
book preview page numper 4
book preview page numper 5
book preview page numper 6
book preview page numper 7
book preview page numper 8
book preview page numper 9
book preview page numper 10
book preview page numper 11
book preview page numper 12
book preview page numper 13
book preview page numper 14
book preview page numper 15
book preview page numper 16
book preview page numper 17
book preview page numper 18
book preview page numper 19
book preview page numper 20
book preview page numper 21
book preview page numper 22
book preview page numper 23
book preview page numper 24
book preview page numper 25
book preview page numper 26
book preview page numper 27
book preview page numper 28
book preview page numper 29
book preview page numper 30
book preview page numper 31
book preview page numper 32
book preview page numper 33
book preview page numper 34
book preview page numper 35
book preview page numper 36
book preview page numper 37
book preview page numper 38
book preview page numper 39
book preview page numper 40
298 pages
Cookie-Einstellungen