Organising knowledge in the age of the semantic web: a study of the commensurability of ontologies

Magee, L 2010, Organising knowledge in the age of the semantic web: a study of the commensurability of ontologies, Doctor of Philosophy (PhD), Global Studies, Social Science and Planning, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Magee.pdf Thesis application/pdf 3.61MB
Title Organising knowledge in the age of the semantic web: a study of the commensurability of ontologies
Author(s) Magee, L
Year 2010
Abstract  This study is directed towards the problem of conceptual translation across different data management systems and formats, with a particular focus on those used in the emerging world of the Semantic Web. Increasingly, organisations have sought to connect information sources and services within and beyond their enterprise boundaries, building upon existing Internet facilities to offer improved research, planning, reporting and management capabilities. The Semantic Web is an ambitious response to this growing demand, offering a standards-based platform for sharing, linking and reasoning with information. The imagined result, a globalised knowledge network formed out of mutually referring data structures termed "ontologies", would make possible new kinds of queries, inferences and amalgamations of information. Such a network, though, is premised upon large numbers of manually drawn links between these ontologies. In practice, establishing these links is a complex translation task requiring considerable time and expertise; invariably, as ontologies and other structured information sources are published, many useful connections are neglected. To combat this, in recent years substantial research has been invested into "ontology matching" - the exploration of algorithmic approaches for automatically translating or aligning ontologies. These approaches, which exploit the explicit semantic properties of individual concepts, have registered impressive precision and recall results against humanly-engineered translations. However they are unable to make use of background cultural information about the overall systems in which those concepts are housed - how those systems are used, for what purpose they were designed, what methodological or theoretical principles underlined their construction, and so on. The present study investigates whether paying attention to these sociological dimensions of electronic knowledge systems could supplement algorithmic approaches in some circumstances. Specifically, it asks whether a holistic notion of commensurability can be useful when aligning or translating between such systems.
     The first half of the study introduces the problem, surveys the literature, and outlines the general approach. It then proposes both a theoretical foundation and a practical framework for assessing commensurability of ontologies and other knowledge systems. Chapter 1 outlines the Semantic Web, ontologies and the problem of conceptual translation, and poses the key research questions. Conceptual translation can be treated as, by turns, a social, philosophical, linguistic or technological problem; Chapter 2 surveys a correspondingly wide range of literature and approaches.
     The methods employed by the study are described in Chapter 3. Chapter 4 critically examines theories of conceptual schemes and commensurability, while Chapter 5 describes the framework itself, comprising a series of specific dimensions, a broad methodological approach, and a means for generating both qualitative and quantitative assessments. The second half of the study then explores the notion of commensurability through several empirical frames. Chapters 6 to 8 applies the framework to a series of case studies. Chapter 6 presents a brief history of knowledge systems, and compares two of these systems - relational databases and Semantic Web ontologies. Chapter 7, in turn, compares several "upper-level" ontologies - reusable schematisations of abstract concepts like Time and Space. Chapter 8 reviews a recent, widely publicised controversy over the standardisation of document formats. This analysis in particular shows how the opaque dry world of technical specifications can reveal the complex network of social dynamics, interests and beliefs which coordinate and motivate them. Collectively, these studies demonstrate the framework is useful in making evident assumptions which motivate the design of different knowledge systems, and further, in assessing the commensurability of those systems. Chapter 9 then presents a further empirical study; here, the framework is implemented as a software system, and pilot tested among a small cohort of researchers. Finally, Chapter 10 summarises the argumentative trajectory of the study as a whole - that, broadly, an elaborated notion of commensurability can tease out important and salient features of translation inscrutable to purely algorithmic methods - and suggests some possibilities for further work.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Global Studies, Social Science and Planning
Keyword(s) Commensurability
Semantic Web
Ontologies
Knowledge Representation
Science and Technology Studies
Versions
Version Filter Type
Access Statistics: 254 Abstract Views, 3502 File Downloads  -  Detailed Statistics
Created: Mon, 27 Aug 2012, 09:30:27 EST by Keely Chapman
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us