## Abstract Radial basis functions are of interest in connection with a variety of approximation problems in the neural networks area, and in other areas as well. Here we show that the members of some interesting families of shift‐varying input–output maps, that take a function space into a functio
Gaussian basis functions for chemometrics
✍ Scribed by Tuomas Kärnä; Francesco Corona; Amaury Lendasse
- Publisher
- John Wiley and Sons
- Year
- 2008
- Tongue
- English
- Weight
- 188 KB
- Volume
- 22
- Category
- Article
- ISSN
- 0886-9383
- DOI
- 10.1002/cem.1166
No coin nor oath required. For personal study only.
✦ Synopsis
Abstract
High‐dimensional data are becoming more and more common, especially in the field of chemometrics. Nevertheless, it is generally known that most of the commonly used prediction models suffer from curse of dimensionality that is the prediction performance degrades as data dimension grows. Therefore it is important to develop methodology for reliable dimensionality reduction. In this paper, we propose a method that is based on functional approximation using Gaussian basis functions. The basis functions are optimised to accurately fit the spectral data using nonlinear Gauss—Newton algorithm. The fitting weights are then used as training data to build a least‐squares support vector machine (LS‐SVM) model. To utilise the reduced data dimension, relevant variables are further selected using forward‐‐backward (FB) selection. The methodology is experimented with three datasets originating from the food industry. The results show that the proposed method can be used for dimensionality reduction without loss of precision. Copyright © 2008 John Wiley & Sons, Ltd.
📜 SIMILAR VOLUMES
Small split-valence Gaussian 3-21SP and 4-22SP basis sets, w Ž .x previously reported for the first-row atoms Chem. Phys. Lett., 229, 151 1996 , have been extended for the second-row elements of the Periodic Table . The total energies of the ground states of the second-row atoms calculated with the
In this paper, the inequality estimates of Bernstein basis functions and Meyer᎐Konig and Zeller basis functions are studied. Exact bounds for these two basis functions are obtained. Moreover, some application results of the new estimates in estimating the rate of convergence of Durrmeyer operators a
The shift operator technique is used for deriving, in a unified manner, the master formulas for the four-center repulsion integrals involving Gaussian (GTO), Slater (STO), and Bessel (BTO) basis functions. Moreover, for the two classes of exponential-type functions (ETO), i.e., STO and BTO, we give
This article investigates a new approach to local approximations to exchange. When a finite basis set is introduced, any operator may be regarded as equivalent to a local operator if its matrix can be reproduced as the matrix of a multiplicative function operator. The expectation values of such oper
Analytical solutions to the Yukawa-like screened Coulomb nuclear attraction and electron repulsion molecular basic integrals, as well as to the basic integrals required to compute the virial coefficient, over Gaussian basis functions, are derived and cast into a practical closed form, suitable to in