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

Improved multidimensional scaling analysis using neural networks with distance-error backpropagation.

L Garrido1, S Gomez, J Roca

  • 1Departament d'Estructura i Constituents de la Matèria,IFAE, Universitat de Barcelona, E-08028 Barcelona, Spain.

Neural Computation
|March 23, 1999
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

Effect of astaxanthin in extenders on sperm quality and functional variables of frozen-thawed boar semen.

Animal reproduction science·2020
Same author

Towards a European health research and innovation cloud (HRIC).

Genome medicine·2020
Same author

Multimorbidity as a predictor of health service utilization in primary care: a registry-based study of the Catalan population.

BMC family practice·2020
Same author

Post-discharge impact and cost-consequence analysis of prehabilitation in high-risk patients undergoing major abdominal surgery: secondary results from a randomised controlled trial.

British journal of anaesthesia·2019
Same author

Boar semen proteomics and sperm preservation.

Theriogenology·2019
Same author

Porcine blastocyst viability and developmental potential is maintained for 48 h of liquid storage at 25 °C without CO<sub>2</sub> gassing.

Theriogenology·2019

Neural networks can perform metric multidimensional scaling (MDS), a dimensional reduction technique. This approach surpasses traditional classical scaling methods in preserving pattern distances.

Area of Science:

  • Machine Learning
  • Data Science
  • Computational Statistics

Background:

  • Metric multidimensional scaling (MDS) is crucial for dimensionality reduction.
  • Classical scaling is the standard algebraic method for metric MDS.
  • Existing methods may have limitations in preserving complex distance relationships.

Purpose of the Study:

  • To investigate the efficacy of neural networks for metric multidimensional scaling.
  • To compare neural network performance against classical scaling.
  • To introduce a novel backpropagation-based approach for MDS.

Main Methods:

  • Utilized neural networks with a specialized error function for backpropagation.
  • Applied the neural network approach to metric multidimensional scaling tasks.

Related Experiment Videos

  • Compared results against the established classical scaling algorithm.

    • Main Results:

    • Neural networks successfully performed metric multidimensional scaling.
    • The neural network approach demonstrated superior performance in preserving original distances compared to classical scaling.
    • The proposed error function enabled effective backpropagation for MDS.

    Conclusions:

    • Neural networks offer a powerful and effective alternative for metric multidimensional scaling.
    • The backpropagation-based neural network method outperforms classical scaling.
    • This research opens new avenues for applying deep learning to dimensionality reduction problems.