“School of Cognitive Sciences”
Back to Papers HomeBack to Papers of School of Cognitive Sciences
| Paper IPM / Cognitive Sciences / 8365 |
|
||||||
| Abstract: | |||||||
|
Hydroxy1 functional group identification has become possible without the use of reagents by using near-infrared (NIR) spectroscopy and artificial neural network (ANN). In this study, the NIR spectra of six components were fed to back propagation three-layer neural network as a training set and then the spectra of 33 chemicals were tested by ANN. The corresponding mathews correlation coefficient for classification task by ANN was calculated. With regard to training time and the error of prediction, a systematic approach was used to determine the number of hidden nodes of ANN. The activation of output nodes and the number of carbons in a series of aliphatic alcohols are highly correlated. Thus, the identification of alcohols with out the use of chemical reagents is also possible by this treatment. The information content of this system (NIR-ANN) for identification of hydroxy1 functional group and the series of aliphatic alcohols was calculated using Shannon's equation
Download TeX format |
|||||||
| back to top | |||||||
