Integration of knowledge-based systems techniques with conventional software techniques in a traditional software development environment

Dai, Wei (1995) Integration of knowledge-based systems techniques with conventional software techniques in a traditional software development environment. PhD thesis, Murdoch University.

Abstract

This thesis investigates how Artificial Intelligence, particularly knowledge-based systems techniques can be integrated with conventional software techniques, and develop an environment in which both knowledge-based systems and conventional software techniques can be utilised equally and easily in the development of intelligent software products. Potential strategies for integrating knowledge-based systems techniques into a traditional Software Development Environments (SOE) have been investigated. These strategies have been used in defining a layered software architecture to support language-level, tool-level and system-level integrations. A taxonomy is developed within the architecture for describing integration ( e.g. to allow the shared computing techniques/resources from different disciplines) and its process at different levels. For knowledge-based systems applications, major limitations of knowledge and inference complexity, and application scale have been improved. The software reuse aspects of knowledge­-based systems development are explored. For software engineering, this work will enable a wider range of scientists to utilise KBS techniques and obtain automated, intelligent support in large-scaled applications. Firstly, the software engineering aspects of the SOE allow the construction of more robust and efficient expert systems. Secondly, the use of knowledge-based systems tools in the environment itself can turn the often rigid and static methodology of software production into a freer and more adaptive process. Thirdly, software tools that use techniques from both paradigms can be easily implemented. The system produced from this research was used to solve both AI and conventional problems, including fault diagnosis in telecommunications, version control and multimedia database applications.

Item Type:	Thesis (PhD)
Murdoch Affiliation(s):	School of Physical Sciences, Engineering and Technology
Notes:	Note to the author: If you would like to make your thesis openly available on Murdoch University Library's Research Repository, please contact: repository@murdoch.edu.au. Thank you.
Supervisor(s):	Wright, Sid
URI:	http://researchrepository.murdoch.edu.au/id/eprint/50251

Item Control Page