Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 35
DEVELOPMENTS IN ARTIFICIAL INTELLIGENCE FOR CIVIL AND STRUCTURAL ENGINEERING Edited by: B.H.V. Topping
Paper I.1
Novel Knowledge Acquisition Techniques for Building Intelligent Information Systems S.M. Griffin* and K. Ahmad#
*INKE Ltd., Surrey Research Park, Guildford, UK
S.M. Griffin, K. Ahmad, "Novel Knowledge Acquisition Techniques for Building Intelligent Information Systems", in B.H.V. Topping, (Editor), "Developments in Artificial Intelligence for Civil and Structural Engineering", Civil-Comp Press, Edinburgh, UK, pp 1-11, 1995. doi:10.4203/ccp.35.1.1
Abstract
The acquisition of knowledge for expert systems is most
frequently impeded by a lack of communication between the
experts and the knowledge engineers involved in building
such systems. This is sometimes rooted in the knowledge
engineer's lack of understanding of the specialist terms of
the domain and at other times this is rooted in the fact that
the knowledge engineer cannot ask the relevant questions of
the expert. We propose a group-oriented knowledge
acquisition methodology where members of the specialist
domain community form an interface group which then helps
in formulating relevant questions of the expert, helps in
elaborating specialist terminology, and generally critiques the
inputs and outputs to and from the knowledge engineer. This
methodology can be fruitfully supplemented by creating a
collection of domain texts. including transcripts of the
interviews and other material suggested by the interface
group, for extracting terms and heuristics. Various aspects
of the proposed methodology are elaborated with reference
to an expert systems development project executed by the
authors over the last three years.
purchase the full-text of this paper (price £20)
go to the next paper |
|