Premium
Choice of Optimum Model Parameters in Artificial Neural Networks and Application to X‐ray Fluorescence Analysis
Author(s) -
Luo Liqiang,
Guo Changlin,
Ma Guangzu,
Ji Ang
Publication year - 1997
Publication title -
x‐ray spectrometry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.447
H-Index - 45
eISSN - 1097-4539
pISSN - 0049-8246
DOI - 10.1002/(sici)1097-4539(199701)26:1<15::aid-xrs182>3.0.co;2-8
Subject(s) - artificial neural network , calibration , computer science , matrix (chemical analysis) , artificial intelligence , biological system , mathematics , statistics , chemistry , chromatography , biology
The model parameters in artificial neural networks have a great influence on the training speed. It can be increased after choosing the optimum parameters, which was performed by a stepping technique. The training speed using the method is usually faster than that when adopting random or empirical parameters. An artificial neural network model was used in multivariate matrix calibration and compared with cross‐validation and partial least‐squares methods, which were combined with the fundamental‐parameters in x‐ray fluorescence analysis. The results show that the artificial neural network model produced the highest accuracy. © 1997 by John Wiley & Sons, Ltd.