Elsevier

Data & Knowledge Engineering

Volume 101, January 2016, Pages 1-23
Data & Knowledge Engineering

An analysis of ontologies and their success factors for application to business

https://doi.org/10.1016/j.datak.2015.11.003Get rights and content

Abstract

Ontologies have been less successful than they could be in large-scale business applications due to a wide variety of interpretations. This leads to confusion, and consequently, people from various research communities use the term with different – sometimes incompatible – meanings. This research work analyzes and clarifies the term ontology and points out its difference from taxonomy. By way of two business case studies, both their potential in ontological engineering and the perceived requirements for ontologies are highlighted, and their misuse in research and business is discussed. In order to examine the case for applying ontologies in a specific domain or use case, the main benefits of using ontologies are defined and categorized as technical-centered or user-centered. Key factors that influence the use of ontologies in business applications are derived and discussed. Finally, the paper offers a recommendation for efficiently applying ontologies, including adequate representation languages and an ontological engineering process supported by reference ontologies. To answer the questions of when ontologies should be used, how they can be used efficiently, and when they should not be used, we propose guidelines for selecting an appropriate model, methodology, and tool set to meet customer requirements while making most efficient use of resources.

Introduction

Computer science and software engineering are relatively recent disciplines compared to other sciences and philosophy. The field continues to evolve as languages mature, representations develop, and the ability to devise solutions to new challenges increases with more advanced software and hardware systems. Artificial intelligence (AI) has been a significant challenge for computer scientists, and while the community has yet to develop “true” human-level machine intelligence, the pursuit of AI has led to the development of knowledge representation and semantic relationships [1], [2]. An ontological representation allows modeling meaning in systems that are to be implemented using a programming language and a database schema. Unlike general software development approaches, such as object-oriented programming models, which enable the transformation of a model into a useful software artifact [3], an ontological model allows software to be generated that can evaluate semantic relationships, validate statements made within a domain of knowledge, and provide much richer rules for information management [4]. As required in artificial intelligence, an ontological model allows known facts and/or assumptions to be used to derive a conclusion or to make inferences (i.e., reasoning).

However, although ontological engineering has been applied for decades, there are still very few truly ontology-based systems that exploit all the benefits of an ontology-based approach (i.e., reasoning) and do much more than classify knowledge into convenient categories. This paper does not aim to provide a broad discussion of the term ontology merely by comparing the semantic differences between several definitions. Rather, the remainder of this paper focuses on the analysis of ontologies and their frequent misuse in research and business and examines why ontologies have not been as successful as they could be in large-scale business applications. Ontological engineering is discussed in detail and compared with both entity-relationship modeling for conceptual database design and software engineering for software design in order to identify their proven benefits and best practices and, whenever possible, to adapt them to ontological engineering.

This paper seeks to clarify the benefits of an ontology-based approach. First, reasons for the frequent misuse or rare use of ontologies in research and business are discussed to examine when ontologies should be used, how they can be used efficiently, and when they should not be used. Guidelines to support the decision for a correct model, methodology, and tool set to meet the project specifications and the customer requirements are proposed. For this purpose, the resources available should be used most efficiently as it is widely established in software engineering to clearly identify which approach to take under given circumstances with a high degree of confidence. Considering the benefits of a clear and machine-interpretable basis for meaning in a system built on ontologies, this is a great opportunity to improve knowledge management and decision making in software systems. Another important benefit of an ontological approach, as discussed further below, is that concepts, their meaning, and their relationships can be shared [5]; this makes possible a clear and unambiguous agreement between a larger number of participants and offers new opportunities in terms of data interchange and data interpretation by machines.

Section 2 provides an overview of ontologies and their background, including a brief discussion of their benefits and a differentiation from taxonomies, and concludes that the general understanding of ontologies is low. This has an impact on their uptake and further deployment, including reuse. In Section 3, case studies from industry and research (i) point out their potential for ontological engineering and their perceived requirements for ontologies and (ii) identify reasons for the misuse of ontologies in research and business. Section 4 discusses in detail the benefits of appropriately used ontologies. Further, it provides a comparative classification of other existing models and technologies, whose relations to ontologies to support the decision for the best-suited methodology or method for each case. Ontological modeling is compared to other conceptual modeling approaches in Section 6, which leads to the question whether ontological engineering is a new development or just a new term for something that already exists. Examining the field of software engineering reveals a number of ontological engineering developments that have trailed software engineering process models. As demonstrated, the software engineering community has already dealt with most of the process models and engineering challenges, and while there are more subtleties and abstractions in the area of ontological engineering, a main goal of this paper is to motivate ontological practitioners to consider adopting some of software engineering’s successful techniques, including the increasing reuse of existing ontologies by applying reference ontologies. Finally, ontological engineering is motivated as a separate and valuable discipline. The discussion concludes with (i) indicators and recommendations for choosing appropriate models and tools and (ii) a critical analysis of ontologies and their application in business in order to improve the maturity and capabilities of ontological engineering. This process must be accompanied by a mature community understanding of what is being discussed and, most importantly, when a project is or is not ontologically based. In order to provide a recommendation for efficient application of ontologies in a (business) project, the following key influencing factors and requirements were identified as being relevant:

  • Requirement for sharing

  • Semantic expressiveness

  • Complexity of the universe of discourse

  • Size of the sharing community (ontology stakeholders)

The most important concepts, their definitions, and meanings used and discussed in this research paper are introduced in Table 1. Based on comprehensive research, we define an ontology by extending the definition by Studer et al. [1] with the key influencing factors which constitute the main benefits of ontologies (cf. Section 4.2.):

An ontology is a formal, explicit specification of a shared conceptualization that is characterized by high semantic expressiveness required for increased complexity.

Section snippets

Information technology following philosophy

The general misuse of ontologies can be explained by (i) many existing, sometimes conflicting, definitions; (ii) insufficiently precise specifications of semantic technologies; and (iii) the existence of complex modeling processes that are too abstract or too complex. Since the term ontology is used with a variety of meanings, some are accurate, it has become increasingly difficult to identify who has or has not using ontologies correctly. The semantic web [11] with its characteristics and

Case studies

In order to further emphasize the importance of target-oriented investment regarding model, representation, and resources, this section introduces two case studies from our portfolio of project collaborations. Both case studies are analyzed in terms of their potential in ontological engineering, and the perceived requirements for ontologies are highlighted considering the following factors:

  • Characteristics of the target domain and the stakeholders involved.

  • Reasons for applying or refraining from

Benefits of ontologies

In order to apply ontologies successfully, they should be used efficiently and for a sound reason. The term ontology is more than a buzzword, and it should certainly be more than an incorrect synonym for taxonomy. The following key questions assist in making the correct decision as to whether ontologies should be used:

  • What are the benefits of ontologies as concepts?

  • Is reasoning their main advantage?

  • What are the benefits gained from applying ontologies?

Benefits cannot be considered solely from a

Guidelines for model selection

Providing guidelines to support the decision for the most appropriate model requires discussing the differences between existing modeling languages and modeling concepts. Several approaches to classifying types of ontologies have been introduced (cf. [16]), and various models have been discussed [70], [71], [72]. Hepp [16] classified ontology projects – but not ontologies – according to the following six characteristics: (i) expressiveness, (ii) size of the relevant community, (iii) conceptual

Ontological engineering supported by reference ontologies

A review of important facts in the context of ontologies reveals that many other modeling approaches, such as entity-relationship modeling [24] and UML [23], also enable analysis, structuring, and organization of a domain of interest [73], [74]. Since in software engineering structured procedures are indispensable for software development, ontological development also requires an appropriate procedure which – provided that a suitable model is selected and correctly employed in the engineering

Conclusion

In this paper, we have clarified the term ontology and have given reasons why ontologies have not been as successful as they could be in large-scale business applications. A critical analysis of existing research and business applications revealed frequent misuse of ontologies. This paper covers four main contributions to overcoming most of the deficits discussed.

  • 1.

    An extended definition of the term ontology and identification of benefits gained from a targeted application of ontological

Dr. Christina Feilmayr is a senior scientist since 2008 at the Institute of Application Oriented Knowledge Processing (FAW), Johannes Kepler University Linz. Her PhD was awarded in 2014 from the Johannes Kepler University and was in the area of information extraction and deals with completing information extraction results via data/text mining methods. Her research interests are in information extraction, natural language processing, text and data mining, and ontology engineering.

References (85)

  • T. Gruber

    A translation approach to portable ontology specifications

    Knowl. Acquis.

    (1993)
  • E. Simperl

    Reusing ontologies on the semantic web: a feasibility study

    Data Knowl. Eng.

    (2009)
  • R. Studer et al.

    Knowledge engineering: principles and methods

    Data Knowl. Eng.

    (1998)
  • S.J. Russel et al.

    Artificial Intelligence—A Modern Approach

  • J. Rumbaugh et al.

    Object-Oriented Modeling and Design

    (1991)
  • N.F. Noy et al.

    Ontology Development 101: A Guide to Creating Your First Ontology

    (2001)
  • M.L. Brodie

    On the development of data models

  • Y. Wand et al.

    An ontological analysis of the relationship construct in conceptual modeling

    ACM Trans. Database Syst.

    (1999)
  • M. Alavi et al.

    Review: knowledge management and knowledge management systems: conceptual foundations and research issues

    MIS Q.

    (2001)
  • R. Davis et al.

    What is a knowledge representation?

    AI Mag.

    (1993)
  • V.C. Storey

    Understanding Semantic Relationships

    (1993)
  • T. Berners-Lee et al.

    The Semantic Web

    Scientific American Magazine

    (March 26 2008)
  • H. Stuckenschmidt

    Ontologien: Konzepte, Technologien und Anwendungen (German)

    (2011)
  • R. Neches et al.

    Enabling Technology for Knowledge Sharing

    AI Mag.

    (1991)
  • N. Guarino et al.

    Ontologies and Knowledge Bases: Towards a Terminological Clarification, Towards Very Large Knowledge Bases: Knowledge Building and Knowledge Sharing

  • W. Borst

    Construction of Engineering Ontologies

    (1997)
  • M. Hepp

    Ontologies: State of the Art, Business Potential, and Grand Challenges, Ontology Management: Semantic Web, Semantic Web Services, and Business Applications

  • C. Mangold

    A survey and classification of semantic search approaches

    Int. J. Metadata Semant. Ontol.

    (2007)
  • D.L. Dong et al.

    Knowledge vault: a web-scale approach to probabilistic knowledge fusio

  • F. Bauer et al.

    Linked Open Data: The Essentials—A Quick Start Guide for Decision Makers

  • O. Bodenreider

    Ontologies and data integration in biomedicine: success stories and challenging issues

  • M. Hepp

    Possible ontologies: how reality constrains the development of relevant ontologies

    Internet Comput.

    (2007)
  • N. Guarino et al.

    What Is an Ontology?

  • Object Management Group OMG

    UML 2.5 Specification

  • P. Chen

    The entity-relationship model—toward a unified view of data

    ACM Trans. Database Syst.

    (1976)
  • M. Hori

    Semantic annotation for web content adaptation

    Spinning the Semantic Web

    (2003)
  • HTML5 Semantic Elements

  • R. Van Rees

    Clarity in the usage of the terms ontology, taxonomy and classification

    CIB Rep.

    (2003)
  • K. Prantner et al.

    Tourism ontology and semantic management system: state-of-the-art analysis

  • G. Dragan et al.

    Model Driven Engineering and Ontology Development

    (2009)
  • P. Lambe

    Organising Knowledge: Taxonomies, Knowledge and Organisational Effectiveness, Chandos

    (2007)
  • T. Leitner et al.

    Optimizing early detection of production faults by applying time series analysis on integrated information

    Int. J. Adv. Syst. Meas.

    (2015)
  • C.J. Janse et al.

    A genotype and phenotype database of genetically modified malaria-parasites

    Trends in Parasitol.

    (2010)
  • M. Uschold et al.

    Ontologies: Principles, Methods and Applications

    (1996)
  • M. Vujasinovic et al.

    Semantic B2B-integration using an ontological message metamodel

    Concurr. Eng.

    (2010)
  • H. Werthner et al.

    E-commerce and tourism

    Mag. Commun. ACM

    (2004)
  • C. Feilmayr et al.

    Covering the semantic space of tourism: an approach based on modularized ontologies

  • Deutscher Tourismusverband: Classification of Tourism Offers

  • Deutsche Hotelklassifizierung

  • ISO 18513:2003

    Tourism Services – Hotels and Other Types of Tourism Accommodation – Terminology

  • World Tourism Organization

    Thesaurus on Tourism & Leisure Activities

    (2001)
  • Open Travel
  • Cited by (0)

    Dr. Christina Feilmayr is a senior scientist since 2008 at the Institute of Application Oriented Knowledge Processing (FAW), Johannes Kepler University Linz. Her PhD was awarded in 2014 from the Johannes Kepler University and was in the area of information extraction and deals with completing information extraction results via data/text mining methods. Her research interests are in information extraction, natural language processing, text and data mining, and ontology engineering.

    Prof. Dr. Wolfram Wöß is an associate professor for applied computer science at the Johannes Kepler University Linz (JKU), Austria, since 2002. Between 1993 and 2002, he was a university assistant at the Institute for Application Oriented Knowledge Processing (JKU Linz, Austria). Wolfram Wöß managed several research and development projects and supervised a number of PhD students.

    His research interests include intelligent information systems, integrated information systems, semantic information integration, ontologies, semantic web, knowledge-centered systems, information engineering, data modeling, data science, data mining, data quality in information systems, business intelligence, and e-business systems.

    View full text