One of the primer objectives of AI is to build systems that are able to enlarge and modify their knowledge base and to profit in an "intelligent" way from the acquired experience. This paper presents a prototype system that provides the capability of adaptively learning from training experience, process that is currently carried out by the user when performing a structural analysis.

The presented system is based on the development of a hierarchy of "intelligent" modules, suited for modeling structures. Emphasis is placed upon the following problems: (1) representation of these modules, (2) the relation between them in the modeling process and the flow of information in the analysis process, (3) defining the concept of training and developing a suited mechanism of training. The system is written in CPROLOG and runs on a VAX 780.

The modular structure of the developed system, as well as the capability to learn from training experience in a dynamic environment, establishes the basis of a formalized methodology for developing training systems which may be extended to include other types of training objectives in the process of structural analysis.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £34 +P&P)