Submitted by: Submitted by ekeen
Views: 110
Words: 7516
Pages: 31
Category: Philosophy and Psychology
Date Submitted: 01/29/2014 07:49 PM
The Relevance of Perspectivism to the Task of Modularisation in Ontology Development
Emily Keen, Dr Simon Milton, Dr Chris Keen
Department of Computing and Information Systems, University of Melbourne
Emily.Keen@bigpond.com, ckeen@unimelb.edu.au, Simon.Milton@unimelb.edu.au,
Abstract
An ontology development methodology seeks to provide developers with established principles, processes, practices, methods and activities for developing ontologies (Gasevic et al., 2009). Diverse methodologies have been published for the development of ontologies, and have evolved, based on the diverse experiences of researchers and practitioners, and the development teams who surveyed the benefits and shortcomings of the available methodologies in order to determine the applicability of methodologies to particular contexts. An evaluation of existing ontology development methodologies has identified that the concept formulation process is not well defined, or based on rigorous processes (Castro et al., 2006; Winters & Tolk, 2009). In order for the validity of the social realism of the actors in a social setting to be captured, the perspectives of each actor needs to be acknowledged and incorporated into the concept formulation process / framework. This paper demonstrates how consideration of perspectivism leads to a meaningful modularisation of the resultant ontology.
Key Words: Perspectivism, ontology development methodology, grounded theory, modularisation of ontologies
INTRODUCTION
The ontology development methodology in this paper adopts a coding process based on grounded theory (Glaser & Strauss, 1967; and Strauss & Corbin, 1997), with term clarification and disambiguation using a reference lexicon, in this case WordNet (Princeton University, 2010). Grounded theory has been proposed as a way of deriving concepts using qualitative analysis, and is applied to the derivation of an ontology using bottom up coding from rich text (Lamp & Milton, 2007; and Urban,...