Knowledge Engineering
Course Objectives
To provide attendees with a disciplined approach to
developing knowledge based systems based on
the CommonKADS methodology. This is achieved by covering
the KBS development process from identifying a feasible application, through
knowledge elicitation by various techniques, up to the production of a
detailed design ready for implementation.
Emphasis is placed on the modelling of expertise. After the course, attendees will:
- know which factors to consider when deciding whether a knowledge based
solution is appropriate;
- know how to perform knowledge elicitation using various techniques;
- know how to produce a knowledge model by following the CommonKADS methodology;
- understand the basis of good implementation decisions;
- know what validation and verification of KBSs entails.
Content
The course presents the life-cycle of a typical KBS project, outlining techniques and methods to assist with
each stage. There is a significant practical element to the course -- attendees
have the opportunity to try out domain, inference and task modelling using a
knowledge engineering workbench, and to practise several knowledge elicitation
techniques. Specific topics include:
- life-cycle models for the development of KBS, including an overview of the
CommonKADS methodology;
-
selecting and scoping a task for knowledge based application;
-
learning and using knowledge acquisition techniques for obtaining domain-specific knowledge from experts or from documents;
-
learning and using CommonKADS techniques for analysing and structuring knowledge;
-
applying a technique based on CommonKADS for designing a KBS;
-
verifying and validating a knowledge base.
This course is not linked to any specific programming tool, although it does
consider how to produce a design which is appropriate to the intended tool.
Duration
3 or 4 days.
Target Audience
The course is intended for knowledge engineers and other technical
specialists who are interested in a methodological approach for developing knowledge based systems, or who are searching for enabling technologies to support the development of knowledge based systems.
Prerequisites
An awareness of popular KBS programming techniques (particularly production rules and objects) is expected. Some knowledge of KBS development tools and a small amount of programming experience is also desirable.
Registration
The course is no longer available. Contact the
School of Informatics
for study opportunities.
|
