Language selection

Government of Canada / Gouvernement du Canada

Search

Constructing a diagnostic knowledge base for building faults

From National Research Council Canada

Alternative titleCréation d'une base de connaissances pour le diagnostic des désordres du bâtiment
Download
  1. (PDF, 1.3 MiB)
AuthorSearch for: 1; Search for: ; Search for:
Affiliation
  1. National Research Council of Canada. NRC Institute for Research in Construction
SponsorSearch for: International Council for Building Research, Studies and Documentation
FormatText, Article
Conference11th International CIB Congress, June 19, 1989, Paris, France
Subjectcomputer programs; artificial intelligence; building failures; building diagnosis; artificial intelligence; knowledge based systems; machine learning; knowledge acquisition
Abstract
A computer system has been developed to build a knowledge base for automatic diagnosis of building faults. This knowledge-acquisition assistant writes rules for diagnosing problems and verifies the logical consistency of the knowledge base as it grows. The system comprises an inductive rule generation module and a rule verification module tied to a rule editor. These three components are used to formulate the rules in the knowledge base. To perform a diagnosis, the accepted rules are used to help diagnose the next case (or expert system shell). Each subsequent diagnostic case is used to add rules and subsequently the rule base is checked for consistency. This type of approach constrains the form of reporting building faults but allows many authors to co-operate in the building of a rule base. The result is a consistent syntax for diagnosis and an expanding rule base for a self-automating building diagnostic system.
Publication date
In
  • Proceedings of the 11th International CIB Congress: 19 June 1989, Paris, France Theme 3, v.1 (1989): 458–465.
LanguageEnglish
Peer reviewedYes
IdentifierIRC-P-1633
NRC numberNRCC 31103
NRC-IRC-3806
NPARC number20375489
Export citationExport as RIS
Report a correctionReport a correction (opens in a new tab)
Record identifiercb374abd-0c1c-41d7-8ba5-20324143e4b3
Record created2012-07-23
Record modified2024-05-10
Date modified: