Dependable Systems and Software Engineering Group(2)

Nine month progress report submitted for continuation towards a PhD

Toward a Canonical Method to Solve Patterns of Ontology Modelling Issues 6 ____________________________________________________________________________________________________________consistent with one another, although here, the sources involved will be kept separately (Noy and Musen, 2000). Reasoning. This activity deals with the study of the inferring capabilities of the produced ontology. Modularization. (Alan Rector)Out of all these aspects of ontology engineering, this report is primarily focussed in the ontology conceptualization task described above, and on the potential opportunities for improvement in the current state of the art. The idea is to study ontology modelling problems at a specific point in the conceptualization process. The first part of the conceptualization phase is to develop a glossary of terms representative of the domain knowledge obtained during the preceding knowledge acquisition phase. At this point, the construction of the model for the ontology starts and it is at this point that ontologists will have to solve different modelling issues to convert the glossary of terms into an ontology model. For example, what terms in the glossary should be modelled as classes? What terms should become properties, property values, or instances? This is the specific step in the conceptualization phase that this research is intended to focus on. The methodologies that address the creation of ontologies from scratch do not provide enough information at the right level of detail about this specific step of the ontology conceptualization phase (Gòmez-Pérez et al., 2004)(Fernandez-Lopez et al., 2002)(Uschold and King, 1995)(Gruninger and Fox, 1995). They look at this activity in broader terms, from a higher level perspective, or from the point of view of what role in the overall ontology engineering lifecycle it plays and what dependencies it has with other engineering activities. Different methodologies provide different levels of detail on how ontology conceptualization should be performed, but none of them discuss in depth the possible modelling problems stated earlier (or their solutions), that ontologists may be faced with, at that specific point of the ontology creation process. Two examples of previous work that examine ontology modelling issues in the context demanded by this research, can be found in (Noy, 2004) and in (Noy and McGuinness, 2001), which in turn, bases part of its rationale on the principles of object-oriented modelling (Rambaugh et al., 1991). Lastly, in the next section it is introduced a sample case scenario of some ontology modelling issues to illustrate the type of problem described here.

Nine month progress report submitted for continuation towards a PhD

Toward a Canonical Method to Solve Patterns of Ontology Modelling Issues 7 ____________________________________________________________________________________________________________3. Work CompletedThis section presents different artifacts developed during the timeframe covered by this report. Section 3.1 contains certain ontology modeling issues in connection to the ReSIST project (ReSIST, 2006) and Section 3.2 provides a summary of the main work items developed while attending the Fourth European Summer School on Ontological Engineering.3.1. The ReSIST ProjectA good amount of work completed during this time, took place in the context of the ReSIST project. ReSIST stands for Resilience and Survivability in Information Society Technology (IST) and it is a Network of Excellence (NoE) project funded under the Sixth Framework Programme of the European Union (ReSIST, 2006). One of the objectives of the ReSIST project is to create a Knowledge Base (KB) application in the domain of resilient computing, partly inspired by the features demonstrated by the semantic web application CS AKTive Space (Glaser et al., 2004)(Shadbolt et al., 2004) and with many of the same requirements. The aim of the ReSIST Knowledge Base (RKB) is to provide an application to the enduser that could serve as a portal to browse and search all type of information in the field of resilient computing: projects, people, institutions, publications, communities of practice, courses, etc. All the information that constitutes the ReSIST KB is stored as Resource Description Framework (RDF) data that is generated in accordance to different ontologies required to equip the KB of semantic capabilities (Manola and Miller, 2004). The mentioned ontologies are provided in the Web Ontology Language (OWL) format (Dean and Schreiber 2004). Further information of the main components and technologies being used to develop the ReSIST KB application can be found in (Millard et al., 2006). 3.1.1. The Ontologies in the ReSIST Knowledge Base A specific aspect that is highly relevant for this report is the different ontologies being used in the ReSIST KB application and the reasons behind them. This section describes such ontologies. The first ontology is the AKT Reference Ontology (AKT, 2002), initially developed for the CS AKTive Space application (Glaser et al., 2004)(Shadbolt et al., 2004). This ontology provides a model to represent people, institutions, projects, publications, research interests, etc. The ontology fits very well the requirements of the ReSIST KB application and is being re-used in its entirety.

Nine month progress report submitted for continuation towards a PhD

Toward a Canonical Method to Solve Patterns of Ontology Modelling Issues 8 ____________________________________________________________________________________________________________The second ontology is the ReSIST Ontology which is still being developed at present. This ontology models the domain of resilient computing. The third ontology is the ReSIST University Curricula Ontology. It models the concept of a university course related …… 此处隐藏：6383字，全部文档内容请下载后查看。喜欢就下载吧 ……