Anthony M. Orme

Complexity metrics for ontology based information

An ontology defines technical terms and shows interrelationships between terms for particular application areas. XML-based standards such as OWL and DAML provide mechanisms to produce XML based ontologies. Ontologies are used in service matching and dynamic web service composition, and are heavily used in bioinformatics and genomics to characterise the structure of living things. Our research focuses on complexity metrics for ontologies. These complexity metrics are compiled from semantic relationships in an ontology. These metrics will help select the best ontologies in several application areas, including bioinformatics and genomics.

A Parallel Methodology for Reduction of Coupling in Distributed Business-to-business E-commerce Transactions

Recently, new standards for business-to-business (B2B) e-commerce transactions, to reduce extended record locking, relaxed standard database transaction properties. In this paper, we provide a parallel methodology employing a mobile/intelligent agent framework to alleviate extended record locking, while adhering to standard database transaction properties. Our methodology provides a minimum 30% reduction of record locking compared to new B2B standards.

Dynamic ontology version control

Ontologies are used today in many application areas. With the use of ontologies in bioinformatics, as well as their use in semantic web technologies, ontology based software has become widely used. This has led to a need for keeping track of different ontology versions [8], as the operation of software will change as the ontologies it uses change. However, existing approaches to ontology versioning have worked on static ontologies. Thus, the ontology version that a software package will use must be chosen prior to running that package. This requires substantial human oversight, and is therefore a major limitation. In this paper, we examine a dynamic approach to ontology versioning that will automatically provide the correct ontology for a software package on-the-fly. We examine a methodology that employs storing different time stamped ontologies in the same file, and we discuss how this methodology can be used on a real ontology.