Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Multivariate calibration with least-squares support vector machines.

Uwe Thissen1, Bülent Ustün, Willem J Melssen

  • 1Laboratory of Analytical Chemistry, University of Nijmegen, Toernooiveld 1, 6525 ED Nijmegen, The Netherlands.

Analytical Chemistry
|May 29, 2004
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Linear Mixed-Effects Models in chemistry: A tutorial.

Analytica chimica acta·2024
Same author

Water quality monitoring based on chemometric analysis of high-resolution phytoplankton data measured with flow cytometry.

Environment international·2022
Same author

Influence of measurement procedure on the use of a handheld NIR spectrophotometer.

Food research international (Ottawa, Ont.)·2022
Same author

Evaluation and comparison of unsupervised methods for the extraction of spatial patterns from mass spectrometry imaging data (MSI).

Scientific reports·2022
Same author

Convolutional neural networks to predict brain tumor grades and Alzheimer's disease with MR spectroscopic imaging data.

PloS one·2022
Same author

A methodological approach to correlate tumor heterogeneity with drug distribution profile in mass spectrometry imaging data.

GigaScience·2020

Least-squares support vector machines (LS-SVMs) offer a novel nonlinear multivariate calibration method for complex chemical analysis. This approach effectively addresses ill-posed problems and spectral variations, providing robust and accurate models.

Area of Science:

  • Chemometrics
  • Machine Learning
  • Spectroscopy

Background:

  • Traditional Support Vector Machines (SVMs) are limited in handling nonlinearities and ill-posed problems.
  • Existing multivariate calibration methods may struggle with complex spectral interferences.
  • Nonlinear multivariate calibration is crucial for accurate chemical analysis in challenging scenarios.

Purpose of the Study:

  • To introduce and evaluate Least-Squares Support Vector Machines (LS-SVMs) as a nonlinear multivariate calibration technique.
  • To demonstrate the application of LS-SVMs in solving ill-posed chemical problems.
  • To assess the performance of LS-SVMs in handling nonlinear spectral interferences, specifically temperature-induced variations.

Main Methods:

  • Least-Squares Support Vector Machines (LS-SVMs) were employed as the primary nonlinear multivariate calibration method.

Related Experiment Videos

  • The method was tested on a known chemical problem involving nonlinear interferences in spectral data.
  • Specific focus was placed on near-infrared (NIR) spectra affected by temperature variations.

    • Main Results:

    • LS-SVMs demonstrated effective nonlinear regression, extending linear capabilities.
    • Model optimization, pruning, and interpretation were successfully performed using LS-SVMs.
    • LS-SVMs exhibited excellent performance, outperforming other approaches on the studied spectral variation problem.
    • Robust models were developed even with nonlinear interferences and spectral variations.

    Conclusions:

    • LS-SVMs are highly promising for solving ill-posed problems in chemometrics and analytical chemistry.
    • The technique provides a robust and efficient alternative for nonlinear multivariate calibration.
    • LS-SVMs offer advantages in model calculation speed and optimization compared to traditional SVMs.