DocumentCode
2261973
Title
Hypercube clustering method: system modeling via modified group method of data handling
Author
Setoodehnia, Ali ; Cheung, John Y. ; Li, Hong
Author_Institution
Dept. of Electr. Eng., Oklahoma Univ., Norman, OK, USA
fYear
1993
fDate
16-18 Aug 1993
Firstpage
422
Abstract
The Hypercube Clustering method is proposed as data partitioning to separate the data set into training and testing sets. This method along with a training procedure is applied to a Modified Group Method of Data Handling (MGMDH) algorithm to identify the characteristics of an input-output of an unknown system and to construct an appropriate mathematical description. Mathematical formulations are presented along with the simulation results
Keywords
hypercube networks; identification; neural nets; nonlinear systems; data handling; data partitioning; hypercube clustering method; modified group method; neural net applications; nonlinear system; system identification; testing sets; training sets; unknown system input-output; Artificial neural networks; Clustering methods; Data handling; Hypercubes; Mathematical model; Modeling; Neural networks; Nonlinear systems; System identification; System testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems, 1993., Proceedings of the 36th Midwest Symposium on
Conference_Location
Detroit, MI
Print_ISBN
0-7803-1760-2
Type
conf
DOI
10.1109/MWSCAS.1993.342998
Filename
342998
Link To Document