Classification of UV–Vis spectroscopic data using principal component analysis and neural network techniques
| Autor: | Kenneth T. V. Grattan, W.J.O. Boyle, N. Benjathapanun |
|---|---|
| Rok vydání: | 1998 |
| Předmět: |
Spectral shape analysis
Absorption spectroscopy Artificial neural network business.industry Computer science Applied Mathematics Pattern recognition Condensed Matter Physics Back propagation neural network Ultraviolet visible spectroscopy Principal component analysis Artificial intelligence Electrical and Electronic Engineering business Absorption (electromagnetic radiation) Instrumentation Second derivative |
| Zdroj: | Measurement. 24:1-7 |
| ISSN: | 0263-2241 |
| DOI: | 10.1016/s0263-2241(98)00020-7 |
| Popis: | This paper presents a comparative study of the use of principal component analysis (PCA) and neural network methods to determine the nature of species present in multicomponent mixtures from ultraviolet–visible (UV–Vis) absorption spectral data. With the use of the PCA technique, absorption spectra with a 316-dimensional space are reduced to six principal components and then classified using the K nearest-neighbor method. By contrast, with the neural network technique, absorption spectra are transformed to represent the spectral shape information by binary encoding segments of the second derivative of the absorption spectra and then classifying using a back propagation neural network algorithm. It is found that the neural network method offers better performance with a higher accuracy than use of the PCA method. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect