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 Video

Updated: Jun 23, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

[Improving hyperspectral matching method through feature-selection/weighting based on SVM].

Yuan-Yuan Wang1, Yun-Hao Chen, Jing Li

  • 1Institute of Resources Technology and Engineering, Beijing Normal University, Beijing 100875, China. wangyuanyuan@ires.cn

Guang Pu Xue Yu Guang Pu Fen Xi = Guang Pu
|May 22, 2009
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Weighted Mean00:57

Weighted Mean

While taking the arithmetic, geometric, or harmonic mean of a sample data set, equal importance is assigned to all the data points. However, all the values may not always be equally important in some data sets. An intrinsic bias might make it more important to give more weightage to specific values over others.
For example, consider the number of goals scored in the matches of a tournament. While computing the average number of goals scored in the tournament, it may be more important to...

You might also read

Related Articles

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

Sort by
Same author

[Experimental study of the eyelid reconstruction in situ with the acellular xenogeneic dermal matrix].

Zhonghua zheng xing wai ke za zhi = Zhonghua zhengxing waike zazhi = Chinese journal of plastic surgery·2007
Same author

[Mutation analysis of GCH1 gene in Chinese patients with dopa responsive dystonia].

Zhonghua yi xue yi chuan xue za zhi = Zhonghua yixue yichuanxue zazhi = Chinese journal of medical genetics·2007
Same author

[Screening and characterization of marine bacteria with antibacterial and cytotoxic activities, and existence of PKS I and NRPS genes in bioactive strains].

Wei sheng wu xue bao = Acta microbiologica Sinica·2007
Same author

[Collateral supply in patients with severe carotid stenosis].

Zhonghua yi xue za zhi·2007
Same author

[Changes of sleep architecture in patients with narcolepsy].

Zhonghua yi xue za zhi·2007
Same author

[Combined anterior and posterior approach for cervical fracture-dislocation with ankylosing spondylitis].

Zhonghua wai ke za zhi [Chinese journal of surgery]·2007
Same journal

The Laser Rangefinder System in Quadrature Modem and Ambiguity Resolution.

Guang pu xue yu guang pu fen xi = Guang pu·2018
Same journal

Improving the Accuracy of Camera-Based Heart Rate Measurement.

Guang pu xue yu guang pu fen xi = Guang pu·2018
Same journal

Determination of Pb, Cr, Cd, and As in Aluminum-Plastic Packaging Materials via Inductively Coupled Plasma-Mass Spectrometry with Microwave Digestion.

Guang pu xue yu guang pu fen xi = Guang pu·2018
Same journal

Study on Molecular Recognition of Cucurbit[6]uril with Oxytetracycline Molecules by Spectroscopic Methods.

Guang pu xue yu guang pu fen xi = Guang pu·2018
Same journal

Preparation and Properties of Novel Polymer Blue Fluorescent Materials.

Guang pu xue yu guang pu fen xi = Guang pu·2018
Same journal

Effect of the Nitrogen Incorporation on the Microstructure and Photoelectric Properties of N Type Nanocrystalline Silicon Thin Films.

Guang pu xue yu guang pu fen xi = Guang pu·2018
See all related articles

Support Vector Machine (SVM) feature selection enhances spectral matching accuracy by 13%-17%. Local weights, derived from SVM, provide the most accurate and stable results for spectral analysis.

Area of Science:

  • Remote Sensing
  • Machine Learning
  • Spectroscopy

Context:

  • Spectral analysis is crucial for identifying materials and classifying land cover.
  • Traditional spectral matching methods can be limited by high dimensionality and noise.
  • Feature selection is vital for optimizing spectral data analysis.

Purpose:

  • To improve spectral matching algorithms by employing feature selection and weighting based on Support Vector Machines (SVM).
  • To evaluate the effectiveness of different weighting strategies (global vs. local) in enhancing spectral matching accuracy.
  • To iteratively refine feature subsets using SVM classification models and multi-objective optimization.

Summary:

  • Feature selection based on SVM iteratively removed less informative features, improving spectral matching.

Related Experiment Videos

Last Updated: Jun 23, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

  • Three scenarios were tested: SVM feature subset only, subset with global SVM weights, and subset with local SVM weights.
  • Local weights, reflecting spectral distribution proximity to the SVM separation plane, yielded the most accurate and stable spectral matching results, increasing accuracy by 13%-17%.

    • Impact:

    • Demonstrates the significant improvement in spectral matching accuracy and stability through SVM-based feature selection.
    • Highlights the superiority of local weighting schemes in capturing spectral variations within classes.
    • Advances spectral analysis techniques by integrating advanced machine learning for improved feature representation and selection.