Quantitative interpretation of spectra can be achieved by using artificial neural networks with multi-layer architecture. Both back-propagation (BP) and radial basis function (RBF) are implemented and tested with raw absorption spectra and normalized spectra of glucose solutions in MATLAB. Simulation results showed partial least square (PLS) method can have better performance with small number of calibration set. However, with increasing size of data set as in cross validation method, RBF and BP have better performance. With optimal spreading factor, RBF can have the same degree of accuracy but significantly faster convergent speed comparing to BP. Normalization scheme can also significantly affect the performance of both RBF and BP.
|Number of pages||2|
|State||Published - 1 Jan 1998|
|Event||Proceedings of the 1998 2nd International Conference on Bioelectromagnetism - Melbourne, Australia|
Duration: 15 Feb 1998 → 18 Feb 1998
|Conference||Proceedings of the 1998 2nd International Conference on Bioelectromagnetism|
|Period||15/02/98 → 18/02/98|