Overview of types of Ontology in the software development process

2013 IEEE Conference on Open Systems (ICOS), December 2 - 4, 2013, Sarawak, Malaysia

1

Overview of types of Ontology in the software development process Rusli Hj. Abdullah , Salfarina Abdullah Marzanah A. Jabar, Mustafa S. Khalefa Department of Information Systems, Department of Information Systems, Faculty of Computer Science and Information Faculty of Computer Science and Information System UPM, University UPM, System UPM, University UPM, Computer Science Department Computer Science Department 43400 Selangor, Serdang, Malaysia 43400 Selangor, Serdang, Malaysia [email protected]; [email protected]; [email protected]; [email protected]

communication between human beings, (ii) to make communication among software systems possible, and (iii) to improve the enterprise system (ES) design and the quality of the software development process. Reference [5] added that a new temporal ontology for a virtual domain that will contain all related domains should be designed. The new domain should be designed using the following procedures as identified by [6] to identify the purpose, to capture, perform the coding, integrate with any existing ontology, perform evaluation and complete the documentation of the ontology. There is agreement with the above in reference [7]. There is a convention in the ontology group that the incorporation of current ontologies is a more valuable way to make a new ontology and thus cut down on the time, price, and attempts required [I]. Ensuring semantic interoperability among different ontologies for incorporation is a key aspect for building ontologies effectively and guaranteeing semantic interoperability across the domain [8]. An ontology of a given domain explains the constituents of truth within that domain in a methodical way, as well as the interaction between these constituents and their kindred to constituents of other domains. Conditions, such as 'domain', 'constituent', 'reality', and 'relation', are themselves ontological terms, as also are terms, such as 'feature', 'object', 'entity', and 'item', as well as 'being' and 'existence'.

ABSTRACT

Ontology is a word that has been the subject of many studies. It is an important concept in computer science to formally represent knowledge in a way that software can process the knowledge and reason about it. The software engineering ontology assists in defining information for the exchange of semantic project information frameworks. Research into ontological issues has been widely active in various areas. This paper presents the origin of ontology research and gives the different definitions of ontology. The paper provides an overview of ontology and its types including the building and design for an enterprise system. Finally the paper provides a systematic review of the subject.

Keywords: Ontology, types of ontology, building, design of ontology, enterprise system

I. INTRODUCTION

There is no straightforward defmition of ontology. References [I, 2] define ontology as having a specific and a clear picture of an abstract world to be presented in a clear simplified view for some purpose whereby vocabularies are extracted from a domain. Having a clear picture may lead to the concepts and their relations being extracted from the real world [3]. Since ontologies use similar shared vocabularies when describing ideas and relations they can also be used as tools in specifying the meaning of terminology systems in a clear defined manner.

II. General level of ontological types

Furthermore, several ontologies form distinct advantage factors. Therefore, it is not surprising that different categories of ontologies are discovered with different focuses. According to the generality stage, Guarino is of the opinion that the following ontology types are available [II] High-level ontologies - explain common ideas, such as space, time, materials and items. They are separate from a particular domain or problem. Their objective is to unify requirements between large

The authors in reference [4] identified three major significant uses of ontologies: (i) to help in

978-1-4799-0285-9/13/$31.00 ©2013 IEEE

83

2013 IEEE Conference on Open Systems (ICOS), December 2 4, 2013, Sarawak, Malaysia -

2

System).ONIONS (Ontological Integration Of Naive Sources). Information ontologies - specify the framework of data archive information, identifying a framework for the consistent storage of information. For example, Information Artifact Ontology (lAO). Knowledge representation ontologies - specify knowledge conceptualizations with an internal structure that exceeds those of the previous ones. They tend to be focused on a description of a particular knowledge use. For example, the Systematised Nomenclature of Medicine Clinical Terms (SNOMED CT) is a clinical ontology.

groups of users like ( concepts, attributes, axioms, etc.). for example Gene Ontology (GO). Domain ontologies - explain the language relevant to a common domain (such us , computer or medicine), through the expertise of the presented ideas of high-level ontologies.for example Ontology Design OntoDesign. Task ontologies - explain the language relevant to a common task or action (such us , development or sales), through the expertise of the presented ideas of high-level ontologies. For example Task Model Ontology (TOM). Application ontologies - explain ideas that belong at the same time to a domain and a process, by means of the expertise of the ideas of domain ontologies and task ontologies. They are generally matched to tasks performed by the sector organizations when performing an activity. For example, Experimental Factor Ontology (EFO) is used to represent a sample variable from gene expression experimental data.

Another possible way of identifying ontologies is according to the characteristics of the real-world problem that is to be exhibited. In this way, lurisica et al. acknowledged the following classes [14]: Static ontologies - explain the features that are available, their characteristics and the connections current among them. This category supposes that the globe is made up of entities, which are blessed with an exclusive and unchangeable identification. For these, we use conditions, such as entity, attribute, or relationship. For example, Biomedical Resource Ontology (BRO). Dynamic ontologies - explain the factors of the patterned world, which can be modified over time. To design these it may be necessary to use limited state machines, Petri nets, etc. 'Process', 'condition', or 'stat conversions' are examples of Ontology Based Data Access (OBDA) Intentional ontologies - explain the factors of the world of inspirations, objectives, values, choices and elections of the engaged providers. Some model terms in these types of ontology are 'element', 'object', 'broker', or 'support'. For example, Phenotypic Quality Ontology (PATO). Social ontologies - explain social factors such as organizational components, netting or interdependences. For this reason they include terms, such as 'actor', 'position', 'part', 'authority', 'responsibility' or 'commitment'. For example , Ontology for Parasite Lifecycle (OPL).

On the other hand, Fensel [12] recognized the following substitute classification: Generic or common-sense ontologies - catch the common information of the world. They provide primary thoughts and ideas for area, time, state, activities, etc., and are valid for a wide range of domains. For example, Generic Ontology Matching and Mapping Mangament (GOMMO). Representational ontologies - are not supposed to be attached to any particular domain. They provide entities without developing what they might signify. Hence, they determine notions, which show information in an object or framework focused approach. For example, Ontology Based Knowledge Represenation for bioinformatics (OBKRB). Domain ontologies catch the information legitimate for a particular kind of domain (for example, Basic Formal Ontology (BFO). Method and task ontologies - the former provides terminologies particular to problem resolution methods, while the latter provides conditions for particular projects. Both provide an affordable perspective as to the understanding of the domain. For example Logic Based Methods for Ontology Mapping (LOG MAP).

III. Ontology design

During the development of a new ontology, style concepts are needed to foster growth and provide a basis for assessment. Reference [18] recognizes five style principles that should guide the enhancement of ontologies. I. Clarity - explanations should be official, complete, aimed and separate from the social or computational perspective. This results in reducing the number of possible understandings of an idea,

Using the type of conceptualization framework, Van Heijst and colleagues recognized the following types of ontology [13]: Terminological ontologies - specify expressions to be used to signify the information of an examined domain. Then try to acquire a specific language relevant to a specified area. An example of this kind would be the ULMS (Universal Medical Language

84

2013 IEEE Conference on Open Systems (ICOS), December 2 - 4, 2013, Sarawak, Malaysia

3

thereby improving the potency of communication between providers.

Ontology

2. Coherence - questions if the ontology is rationally constant. "If a phrase that can be deduced from the axioms opposes a meaning or example given informally, then the ontology is incoherent." Only implications consistent with the current explanations should be permitted.

, Capturing

Figure

Extendability - ontologies should be developed in a way that allows for the meaning of new conditions for unique uses without requiring change in the current conditions. Moreover the ontology should be developed to work as a conceptual base for a variety of expected projects. It should be possible to expand the ontology without changing the current explanations. If this is the situation, it is not necessary to have a vocabulary sufficient to show the information relevant to all anticipated additions since the systems to determine the need of specialization are already in place. 3.

I

"

,

Coding

T ntegrating with existing ontologies

1. Ontology building stages.

Capture

In this stage, the process of identifying key concepts and relationships in the domain is provided. Furthermore, producing precise and explicit content definitions is mandatory for each model that can easily refer back to the same model. Model identification is a challenging task in the development of any ontology model and there is no general procedure that can be applied for all models. The databases that are used for any ontology model are qualitative. For this reason, model identification is necessary in order to analyze this qualitative data. Therefore, a qualitative analysis tool (NVivo), can be used for analyzing the initial concepts and relationships of the ontology model.

4. Minimum Encoding Bias - ontologies should be developed at the "knowledge level" rather than obligating the ontology to a particular execution language and its particular restnctlOns. Representation options should not be dependent on the comfort of a particular notation or implementation issues at the symbol-level.

Coding

After the model identification of ontology, coding is an essential stage, which involves an explicit representation of the ontology concept. The main purpose of coding is to control the possible misinterpretations of the key concept. Moreover, concepts are based on a hierarchical architecture such as class-superclass and instance-class [19]: There are some typical representations available for ontology documentation, such as Web Ontology Language (OWL), KIF, Cyc, Ontolingua and FLogic [24]. In reference [21], the authors applied OWL for their ontology representation. OWL coding provides multiple purposes of developed disambiguation, relationship structuring, version management and a foundation.

Minimal Ontological Commitment - ontologies should create as few statements as possible about the domain being patterned without losing the functionality of the ontology. Accordingly, the most fragile concept necessary to facilitate the communication consistent the with conceptualization should be specified. This allows providers to quickly increase the ontology where required for their personal requirements. 5.

IV. Ontology Building

Three stages for building an ontology enterprise are recommended by references [21,22],as shown in Figure 1.

Integration with Existing Ontologies

The integration of ontologies is the process of developing new ontologies based on one or more existing relevant ontologies [23]. However, the integration of existing models typically involves aggregating, combining and assembling together

85


4

the source ontologies with the newly developed ontology, extension, specialization and adaptation [25]. To date, no existing ontology has been totally built without depending on a relevant earlier ontology. Furthermore, researchers are still encouraged to develop existing ontologies for better performance and to increase efficiency.

Ontology Representation Languages

Ontology representation languages are of six types: I.

2. 3. 4.

5. 6.

In Dorma I - natura I Ianguage. Semi-Formal - limited structured form of natural language. Formal - official terminology with official semantics. Cycle - designed in the cycle venture. It is dependent on first-order predicate calculus with some higher-order additions. RIF (Rule Interchange Format) and F-Logic - to unite the ontology and the rules OWL - designed as an extension from RDF and RDFS, and previous ontology language projects: OIL, DAML, DAML+OIL. OWL is able to be used over the World Wide Web,

� �

Enterprise Resource Planning system (ERP), Supply Chain enabled Management system (CRM). The main role of these systems is to enable a company to make full use of the resources available, as well as to provide services and/or products to clients through the full functionality and integrated solutions for the company's information processing needs. The most important [22, 25] aspects of ERP are the capabilities to automate and assimilate an organization's business processes, share common as %1to and practices �hrougho.ut t�e enterpr.ise, produce and access IllformatlOn III a real-time environment. This is clarified in Figure (3).

Manufactur� DS

� [ ark eting m.1l : :0i )ll i fOf ��{b iYl

MIS

�!anufactur� ES

Real world

system

s a representation of

Figure 3. Definition of a General Enterprise System

Figure 2. Ontology as a filter of knowledge

Conclusion

This paper is a systematic research review of several ideas concerning ontology types and how to build an ontology. Additionally a description of the main components involved in ontological design has been given including a brief overview of enterprise systems and ontology languages.

Enterprise system

A general name for a packaged application is known as an enterprise system and it is used by enterprises to process transactions and manage the [23] information system for business processes and functions throughout an organization. Enterprise systems are used across the world in different organizations regardless of their size. Some of the most common enterprise systems are a Management system (SCM),

Ontologies are expected to increasingly appear in various areas as promising tools to improve communication among people and to achieve interoperability among systems. They also act as improving agents for humans or software by

86


5

Architecture for Integration of Ontologies" The Semantic Web - ASWC 2006 Lecture Notes in Computer Science Volume 4185, 2006, pp 205-211

reusing a known data model or knowledge. All these tasks deal with interoperability issues and can be applied in different domains.

[9] Johansson, Ingvar 2004 "Ontological Investigations. An Inquiry into the Categories of Nature, Man and Society", New York and London: Routledge.

[10] Frank, A. U., Palmer, 8., and Robinson, V., 2000. "Formal Methods for Accurate Definition of Some Fundamental Terms in Physical Geography". In Proceedings, Second International Symposium on Spatial Data Handling, Williamsville, NY:

REFERENCES

[I] LindaAnticoli, Elio Toppano" "The Role of Culture" in Collaborative Ontology Design International Conference, 2011 - dl.acm.org [2] PISANELLI, D.M., GANGEMI, A. and STEVE, G., "Ontologies and Information Systems: The Marriage of the Century?", Proceedings of the LYEE Workshop, Paris, 2002.

[II] Guarino, N.: "Formal Ontology in Information Systems". In Proceedings of FOIS'98, Trento, Italy. lOS Press, Amsterdam (2002).

[3] GRUBER, T., "A Translation Approach to Portable Ontology Specifications", Knowledge Acquisition. Volume 58, Issue 1, January 2003, Pages 89-123

[12] Fensel, D: "Ontologies: A Silver Bullet for Knowledge Management and Electronic Commerce". Second Edition, Springer Verlag, Berlin, Heidelberg (2004).

[4] GUARINO, N., "Formal Ontology in Information Systems", Proceedings of FOIS'98, Formal Ontology in Information Systems, Handbook on Ontologies, Information Systems, Springer-Verlag Berlin Heidelberg 2009 Page 2. 202 N. Guarino and CA Welty

[13] Van Heijst, G., Schereiber, A.T., Wielinga, B.J.: "Conceptual Modelling for Knowledge Based System" (200I)

[5] Uschold, M., and R. Jasper (1999). "A framework for understanding and classifying ontology applications". In Benjamins et aI., pp. 11-1 - 11-12.

[14] Wouters, B., Deridder, D., Van Paesschen, E.: "The use of ontologies as a backbone for use case management". EuropeanConference on Object-Oriented Programming (ECOOP), 2000.

[6] Y. Kalfoglou and M. Schorlemmer. "Information-flow-based ontology mapping". In On the Move to Meaningful Internet Systems 2002: CoopIS, DOA, and ODBASE, volume 2519 of Lecture Notes in Computer Science, pages 1132-1151. Springer, 2002.

[15] Uschold, M.: "Building Ontologies:Towards a Unified Methodology". In: 16th Annual Conf. of the British Computer Society Specialist Group on Expert Systems (2008)

[7]

[8]

[16] Burton-Jones, A., Storey, V.C., Sugumaran, V., Ahluwalia, P.: "A Semiotic Metrics Suite for Assessing the Quality of Ontologies". Data and Knowledge Engineering 55, 84102 (2005)

Asunci6n G6mez-Perez, Mariano Fernandez L6pez, Oscar Corcho - 2004 "Ontological Engineering: With Examples from the Areas of Knowledge" Jeongsoo Lee, Heekwon Chae, Kwangsoo Kim, Cheol-Han Kim "An Ontology

87


6

"Systems success assessment: Viewpoints of Is function staff and end-users", Industrial Engineering and Engineering Management,. IEEE International Conference.

[17] Booch, G., Rumbaugh, J., Jacobson, I.: "Unified Modeling Language User Guide". Addison-Wesley Professional (2005)

[18] Pinto, H.S., Martins, J.P.: "Ontologies: How can they be Built?", Knowledge and Information Systems 6, 441--464 (2004)

[19] Pulido, J.R.G., Ruiz, M.A.G., Herrera, R., Cabello, E., Legrand, S., Elliman, D.: "Ontology Languages for the Semantic Web: A Never Completely Updated Review". Knowledge-Based Systems 19, 489-497 (2006)

[20] W3C OWL Working Group: OWL 2 Web "Ontology Language - Document Overview". World Wide Web Consortium (2009). [21] M. Barathi and S.Valli, "Ontology Based Query Expansion Using Word Sense Disambiguation." (IJCSIS) International Journal of Computer Science and Information Security, Vol. 7, No. 2, February 2010.

[22] F. Nah, J. Lau, and J. Kuang, "Critical factors for Successful Implementation of Enterprise Systems," Business Process Management Journal, vol. 7 no. 3, 2001, pp. .

[23] D. Sedera, G. Gable, and T. Chan, "Measuring enterprise systems success: The importance of a multiple stakeholder perspective," in Proc. of the 12th European Conference on Information Systems, The European IS Profession in the Global Networking Environment, 14.-16.6.(2004), Turku, Finland, pp. 1-13.

[24] P. E. Ifinedo, "Prioritization of Enterprise Resource Planning (ERP) Systems Success Measures: Viewpoints of Two Organizational Stakeholder Groups," in Proc. 21st Annual ACM Symposium on Applied Computing, SAC'06, Dijon, France, pp. 1554-1560.

[25] P. E. Ifinedo, Kaataja, M.J. Enterprise

88