This paper is an attempt at providing a fuzzy set formalization of case-based reasoning and decision. Learning aspects are not considered here. The proposed approach assumes a principle stating that ''the more similar are the problem description attributes, the more similar are the outcome attribute
Issues and applications of case-based reasoning in design
β Scribed by Underwood, Peter G.
- Publisher
- John Wiley and Sons
- Year
- 1998
- Tongue
- English
- Weight
- 34 KB
- Volume
- 49
- Category
- Article
- ISSN
- 0002-8231
No coin nor oath required. For personal study only.
β¦ Synopsis
inform, aid, and enrich the decisions of designers rather than substitute for their creative talents. For example, Tsatsoulis and Alexander describe (p. 272) a design system to aid expert fire-Anyone occupied with the task of design knows how difficult fighters to contribute to the design of fire engines: They explain it can be to ensure that a new system truly reflects the needs that within the design process, the firefighters may not necessarof the users rather than the preconceptions and preoccupations ily have expertise but they have a vital and unique view of the of the designer. In many cases, the true nature, rich description, functionality of the finished apparatus. An important aspect of and hidden exceptions which characterize systems of even modthis view is that considerations other than the utilitarian may erate complexity are not captured by those formal techniques guide choice of design: In other words, the end result has to which rely upon defining ''rules'' or the construction of a model look good and appeal to clients, as well as working as intended. of a problem domain from first principles. The result is a design which, while it may be elegant and a tribute to the ingenuity
The book consists of 12 chapters, bringing together a wideof the designer, may fall short of expectations and may need a ranging set of papers by experts in design research: The authors lot of fine-tuning before it can play a useful role.
were selected from participants in the Workshop on Case-Based ''Case-based'' reasoning (CBR) offers another approach,
Reasoning in Design at the 1994 International Conference on based upon the experiences of designers and users salient to the Artificial Intelligence in Design. The preliminaries include the system to be designed. It is a relatively recent development, contact address, including E-mail, for each author. Each chapter growing out of research in knowledge engineering. As the edirepresents an expansion by the authors of the original notes tors explain (p. 2), case-based reasoning uses knowledge gained from the Workshop. The aim of the editors, as stated in the from solving similar problems, and concentrates on provision Preface, has been to assemble a significant case library with the of a system which: specific purpose of allowing design researchers to describe their experiences. They have achieved this admirably, and, in so j Finds (retrieves) those cases in a case base that solved probdoing, have produced a book which is enjoyable to read and lems similar to the current problem;
π SIMILAR VOLUMES
Future generations of databases will store increasing amounts of pictorial and other visuospatial material. Among other methods, these will require retrieval techniques which exploit memory for the location of objects in pictures stored within the database. Experiments are reported in which subjects