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

Dynamic characterization of cluster structures for robust and inductive support vector clustering.

Jaewook Lee1, Daewon Lee

  • 1Department of Industrial and Management Engineering, Pohang University of Science and Technology, Kyungbuk, Korea. jaewookl@postech.ac.kr

IEEE Transactions on Pattern Analysis and Machine Intelligence
|October 27, 2006
PubMed
Summary

This study introduces a new topological and dynamical framework for support vector clustering. The developed method enhances inductive clustering by decomposing clusters and utilizing a weighted graph for robustness.

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

Potentiated Maturation of hPSC-Derived Dopaminergic Neurons via Convergent Genetic and Small Molecule Modulation.

Tissue engineering and regenerative medicine·2026
Same author

Atomic Evolution of Hydrogen Intercalation Wave Dynamics in Palladium Nanocrystals Revealed by Liquid-Phase Transmission Electron Microscopy.

Journal of the American Chemical Society·2026
Same author

Lower extremity muscle activation patterns during running in individuals with and without anterior cruciate ligament reconstruction.

Clinical biomechanics (Bristol, Avon)·2026
Same author

Nanocrystal Geometry Governs Phase Transformation Pathways in Palladium Hydride.

ACS nano·2026
Same author

Protocol for formulation and evaluation of phytosterol-based lipid nanoparticles as cholesterol alternatives for mRNA delivery.

STAR protocols·2026
Same author

Retinoic acid signaling regulates astrocyte reactivity by modulating MAPK/NF-κB pathways and mitochondrial integrity.

Neurochemistry international·2026

Area of Science:

  • Data Science
  • Machine Learning
  • Computational Topology

Background:

  • Support vector clustering (SVC) is a powerful technique for data analysis.
  • Understanding the underlying topological and dynamical properties of SVC is crucial for improving its capabilities.
  • Existing methods may lack robustness and inductive learning potential.

Purpose of the Study:

  • To develop a topological and dynamical characterization of support vector clustering.
  • To introduce a novel approach for inductive clustering based on cluster decomposition.
  • To create a robust and effective clustering algorithm.

Main Methods:

  • Decomposition of clusters into basin level cells.
  • Extension of clusters to enlarged clustered domains for inductive learning.

Related Experiment Videos

  • Construction of a simplified weighted graph preserving cluster topology.

    • Main Results:

    • A formal topological and dynamical characterization of SVC is established.
    • The proposed method demonstrates enhanced inductive clustering capabilities.
    • Simulation results validate the robustness and effectiveness of the new algorithm.

    Conclusions:

    • The developed framework provides a deeper understanding of support vector clustering.
    • The proposed inductive clustering algorithm is robust and effective.
    • This work offers a foundation for further advancements in clustering methodologies.