DocumentCode
2913100
Title
Application of the Graph Clustering Algorithm to Analog Systems Diagnostics
Author
Bilski, Piotr
Author_Institution
Warsaw Agric. Univ., Warsaw
fYear
2007
fDate
1-3 May 2007
Firstpage
1
Lastpage
6
Abstract
The paper presents the method for analysis of the learning data sets, used to create automated diagnostic modules. Graph clustering algorithm is presented and applied to the detection of the similarity between the learning examples. Possible applications of the method to the alternative fault codes labeling, ambiguity groups detection, and optimization of the existing diagnostic modules are considered. Experiments using electric machine model are presented and conclusions drawn.
Keywords
graph theory; learning (artificial intelligence); ambiguity groups detection; analog systems diagnostics; automated diagnostic modules; diagnostic module optimization; fault codes labeling; graph clustering algorithm; learning data sets; Artificial intelligence; Clustering algorithms; DC motors; Electrical fault detection; Face detection; Fault detection; Fuzzy logic; Learning; Rough sets; System testing; analog systems; data exploration; diagnostics; machine learning;
fLanguage
English
Publisher
ieee
Conference_Titel
Instrumentation and Measurement Technology Conference Proceedings, 2007. IMTC 2007. IEEE
Conference_Location
Warsaw
ISSN
1091-5281
Print_ISBN
1-4244-0588-2
Type
conf
DOI
10.1109/IMTC.2007.379088
Filename
4258350
Link To Document