Applying KDD techniques to produce diagnostic rules for dynamic systems

Derek Sleeman, F Mitchell, R Milne

Research output: Working paperDiscussion paper

Abstract

This paper argues that there is a discernible trend in Knowledge Acquisition towards systems which are easier for the domain expert to use; such systems ask more focused questions and questions at a higher level. The TIGON system which illustrates this trend is described in some detail. So far TIGON has been used to infer fault detection and fault diagnosis rules for a gas turbine engine. But we argue that the methodology evolved here should make it suitable for use with a wide range of dynamic systems, including scientific ones. The relationship of TIGON to data mining is discussed as is the relationship of this system to other trends in KA, namely:

Co-operative systems for Knowledge Acquisition/Problem Solving
The re-use of existing knowledge(bases)
Original languageEnglish
Place of PublicationAberdeen University
Pages1-13
Number of pages14
VolumeAUCS
Publication statusPublished - Aug 1996

Fingerprint Dive into the research topics of 'Applying KDD techniques to produce diagnostic rules for dynamic systems'. Together they form a unique fingerprint.

  • Cite this

    Sleeman, D., Mitchell, F., & Milne, R. (1996). Applying KDD techniques to produce diagnostic rules for dynamic systems. (TR9604 ed.) (pp. 1-13). Aberdeen University.