1993
DOI: 10.1016/0954-1810(93)90006-2
|View full text |Cite
|
Sign up to set email alerts
|

The development of Bridger: A methodological study of research on the use of machine learning in design

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
19
0

Year Published

1993
1993
2007
2007

Publication Types

Select...
5
2

Relationship

2
5

Authors

Journals

citations
Cited by 23 publications
(19 citation statements)
references
References 31 publications
0
19
0
Order By: Relevance
“…Following a bridge adaptation to suite a new specification, Bridger performed strength analysis according to American design codes. Bridges that failed the test could be modified using an interactive redesign system based on an enhanced version of Protos (Reich 1993a;Reich and Fenves 1995). This therefore is an interesting vertical integration where one CBR system serves to enhance one CBR step of another CBR system.…”
Section: Knowledge Used For Case Revision-cbr Stepmentioning
confidence: 98%
See 2 more Smart Citations
“…Following a bridge adaptation to suite a new specification, Bridger performed strength analysis according to American design codes. Bridges that failed the test could be modified using an interactive redesign system based on an enhanced version of Protos (Reich 1993a;Reich and Fenves 1995). This therefore is an interesting vertical integration where one CBR system serves to enhance one CBR step of another CBR system.…”
Section: Knowledge Used For Case Revision-cbr Stepmentioning
confidence: 98%
“…One example is design and an exemplar of this problem type is Bridger-a system for designing cable-stayed bridges (Reich 1993a;Reich and Fenves 1995). Bridger is capable of creating an elaborate set of design attributes from a minimal set of design specifications.…”
mentioning
confidence: 99%
See 1 more Smart Citation
“…It is unnecessary to assess the relevance of this selection by testing the learning program in the context of the original design problem. COMMENTS/CRITICISM: In most cases, programs are selected because they are readily available and not because they are actually the best for the task [20]. In many cases, the representation "supposedly natural'* to the domain, is selected to fit the learning program available.…”
Section: Assumptionsmentioning
confidence: 99%
“…Many CAD researchers turned to building systems that can automatically learn to cope with this ill-structured problem. Machine learning techniques have been widely adopted in knowledge-based systems to provide knowledge acquisition, modification and generalization, for example ECOBWEB (Reich and Fenves 1991) and BRIDGER (Reich 1993). These systems treat knowledge as universally applicable context-free generalizations and descriptions (Reffat and Gero 2000), so that they can be reused in different circumstances.…”
Section: Introductionmentioning
confidence: 99%