Use of Adaptive Methods to Solve the Inverse Problem of Determination of Composition of Multi-Component Solutions
- Authors: Efitorov A.1, Dolenko S.1, Dolenko T.1,2, Laptinskiy K.1,2, Burikov S.1,2
-
Affiliations:
- Skobeltsyn Institute of Nuclear Physics
- Phaculty of Physics
- Issue: Vol 27, No 2 (2018)
- Pages: 89-99
- Section: Article
- URL: https://journals.rcsi.science/1060-992X/article/view/195075
- DOI: https://doi.org/10.3103/S1060992X18020042
- ID: 195075
Cite item
Abstract
This study considers solving the inverse problem of determination of salt or ionic composition of multi-component solutions of inorganic salts by their Raman spectra using artificial neural networks. From the point of view of data analysis, one of the key problems here is high input dimensionality of the data, as the spectrum is usually recorded in 1–2 thousand channels. The two main approaches used for dimensionality reduction are feature selection and feature extraction. In this paper, three feature extraction methods are compared: channel aggregation, principal component analysis, and discrete wavelet transformation. It is demonstrated that for neural network solution of the inverse problem of determination of salt composition, the best results are provided by channel aggregation.
About the authors
A. Efitorov
Skobeltsyn Institute of Nuclear Physics
Email: dolenko@srd.sinp.msu.ru
Russian Federation, Moscow, 119991
S. Dolenko
Skobeltsyn Institute of Nuclear Physics
Author for correspondence.
Email: dolenko@srd.sinp.msu.ru
Russian Federation, Moscow, 119991
T. Dolenko
Skobeltsyn Institute of Nuclear Physics; Phaculty of Physics
Email: dolenko@srd.sinp.msu.ru
Russian Federation, Moscow, 119991; Moscow, 119991
K. Laptinskiy
Skobeltsyn Institute of Nuclear Physics; Phaculty of Physics
Email: dolenko@srd.sinp.msu.ru
Russian Federation, Moscow, 119991; Moscow, 119991
S. Burikov
Skobeltsyn Institute of Nuclear Physics; Phaculty of Physics
Email: dolenko@srd.sinp.msu.ru
Russian Federation, Moscow, 119991; Moscow, 119991