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 Optimized 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. Toutilise 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.

Department(s)

Engineering Management and Systems Engineering

Publication Status

Full Access

Keywords and Phrases

Dimensionality reduction; LS-SVM; Nonlinearity; Regression; Variable selection

International Standard Serial Number (ISSN)

1099-128X; 0886-9383

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 Wiley, All rights reserved.

Publication Date

01 Jan 2008

Share

 
COinS