The CommonKADS Methodology

Andre Valente and John Kingston

Course Description

When knowledge is acquired for the construction of a knowledge based system (KBS), it must be represented and structured in such a way that it can be used to analyse existing structures and approaches, and to produce a coherent design for a knowledge based system. The separation of analysis and design is a key element in both accurate modelling of expert knowledge in an implementation-independent manner (allowing the possibility of reusing knowledge models between applications) and in making well-justified decisions about the best implementation approach for a particular problem. The CommonKADS methodology provides a principled approach to representing, organising and transforming acquired knowledge; in other words, it is an enabling technology to support and guide the construction of knowledge based systems.

The goal of the tutorial is to introduce the audience to a disciplined approach to developing knowledge based systems based on the CommonKADS methodology. The course is intended for knowledge engineers and other technical specialists interested in methods for developing knowledge based systems. On completion, the audience will understand the benefits of modelling knowledge as an intermediate step between knowledge acquisition and implementation; know how to produce a set of knowledge models by following the CommonKADS methodology; be able to identify the task types of potential and actual KBS applications; and understand the basis of good design decisions.

Prerequisite Knowledge

An awareness of popular knowledge representation and inferencing techniques is expected. Some knowledge of KBS development tools and a small amount of programming experience is also desirable.

About the Lecturers

Andre Valente is a computer scientist at the USC Information Sciences Institute. Prior to his work at ISI he participated, at the University of Amsterdam, in the design of the Common KADS methodology and the CommonKADS Library in particular. He has also worked for major industrial corporations, performing applied research and development in knowledge engineering. He holds a Ph.D. (Computer Science) from the University of Amsterdam (1995). His research interests are knowledge engineering, knowledge acquisition, and planning.

John Kingston is a Senior Computer Scientist within the Knowledge Engineering Methods Group at the Artificial Intelligence Applications Institute (AIAI), University of Edinburgh. He has developed several commercial knowledge based systems, presented a range of commercial training courses, and provided consultancy on knowledge based systems in the UK, Europe and the USA. He also publishes frequently, and was awarded first prize for Best Technical Paper at the BCS Expert Systems '93 conference for a paper on applying the CommonKADS methodology.

Last modified: Thu Feb 20 14:00:43 JST 1997