Multicomponent Kinetic Determination Using an Artificial Neural Network with Maximum Likelihood Principal Component Analysis

Author

Gao, Ling ; Ren, Shouxin

Author_Institution

Dept. of Chem., Inner Mongolia Univ., Huhhot

Volume

3

fYear

2008

fDate

18-20 Oct. 2008

Firstpage

469

Lastpage

473

Abstract

The multilayer feedforward network (MLFN) based on the back propagation (BP) and the Levenberg-Marquardt (LM) algorithm with maximum likelihood principal component analysis (MLPCA) was developed to improve multicomponent kinetic determination. This proposed MLPCA-LM-BP-MLFN method was tested for simultaneous multicomponent kinetic determination of Co(II), Ni(II) and Cu(II) and revealed significantly improved performance over the existing three other methods.

Keywords

backpropagation; biology computing; feedforward neural nets; maximum likelihood estimation; principal component analysis; Levenberg-Marquardt algorithm; MLPCA; artificial neural network; back propagation; maximum likelihood principal component analysis; multicomponent kinetic determination; multilayer feedforward network; Arithmetic; Artificial neural networks; Chemistry; Computer networks; Electronic mail; Jacobian matrices; Kinetic theory; Multi-layer neural network; Principal component analysis; Testing; Artificial Neural Network; Maximum Likelihood Principal Component Analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2008. ICNC '08. Fourth International Conference on

Conference_Location

Jinan

Print_ISBN

978-0-7695-3304-9

Type

conf

DOI

10.1109/ICNC.2008.12

Filename

4667183