Combined use of partial least squares regression and neural network for diagnosis tasks

Proceedings - International Conference on Pattern Recognition

Volumn 4, Issue , 2004, Pages 573-576

  Debiolles, Alexandra ^a   Oukhellou, Latifa ^b   Aknin, Patrice ^c  

^a SNCF   (France)

^b UNIVERSITÉ PARIS EST CRÉTEIL   (France)

^c Lab des Technologies Nouvelles   (France)

Author keywords

[No Author keywords available]

Indexed keywords

FEATURE SELECTION; HIDDEN NODES; OPTIMAL NUMBER; PARTIAL LEAST SQUARES (PLS);

ALGORITHMS; LEAST SQUARES APPROXIMATIONS; PRINCIPAL COMPONENT ANALYSIS; RAILROAD TRACKS; REGRESSION ANALYSIS; VECTORS; VEHICLE TRANSMISSIONS;

NEURAL NETWORKS;

EID: 10044299329     PISSN: 10514651     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/ICPR.2004.1333837     Document Type: Conference Paper

References (5)

1. C.M. Bishop, Neural networks for pattern recognition, Oxford Univ. Press, United States, 1997
2. G. Cron, B. Dubuisson, "Fault detection for time dependent states using fuzzy regression. Application to the French telephone network management", NAFIPS'99, New York City, USA, 1999
3. M. Tenenhaus, La régression PLS théorie et pratique, Technip, France, 1998.
4. T.C.R Hsiao, C.W. Lin, H.K. Chiang, "Partial least-squares algorithm for weights initialization of backpropagation network", Neurocomputing 50, pp. 237-247, 2003
5. L. Oukhellou, P. Aknin, F. Vilette, "Classification and localization of defects by combined use of internal and external modelings", PSIP'03, Grenoble, France, 2003.