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

Automated variable weighting in k-means type clustering.

Joshua Zhexue Huang1, Michael K Ng, Hongqiang Rong

  • 1E-Business Technology Institute, The University of Hong Kong, Pokfulam Road, Hong Kong. jhuang@eti.hku.hk

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 7, 2005
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

Glucose Modulated-Frustrated Lewis Acid-Base Pairs and Oxygen Vacancy for Enhanced Photocatalytic CO<sub>2</sub> Reduction.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

DFIR-DETR: Frequency-domain iterative refinement and dynamic feature aggregation for small object detection.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

Ecosystem simulation: the software to platform leap.

Scientific reports·2026
Same author

SpecEStop: Self-Supervised Hyperspectral Mixed Noise Removal via Deep Spectral Prior.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

The actual Co(10-12) surface structure and CO activation.

Nature communications·2026
Same author

Hierarchical in situ biomineralization-enhanced electroactive biofilm for high-efficiency antibiotic degradation.

Bioresource technology·2026

This study introduces a novel k-means clustering algorithm that automatically calculates variable weights, improving cluster recovery. This method enhances data mining by identifying important variables for complex datasets.

Area of Science:

  • Computer Science
  • Data Mining
  • Machine Learning

Background:

  • Standard k-means clustering assigns equal importance to all variables.
  • Variable selection is crucial for effective clustering in complex, high-dimensional datasets.
  • Existing methods often require manual feature weighting or selection.

Purpose of the Study:

  • To propose a k-means type clustering algorithm with automatic variable weight calculation.
  • To introduce a method for iteratively updating variable weights based on data partitioning.
  • To provide a theoretical guarantee of convergence for the proposed algorithm.

Main Methods:

  • A novel iterative step is integrated into the k-means algorithm.
  • Variable weights are updated based on the current data partition.

Related Experiment Videos

  • A specific formula for calculating variable weights is proposed and analyzed.
  • Convergence of the enhanced clustering process is proven theoretically.

    • Main Results:

    • The algorithm successfully calculates variable weights reflecting their importance in clustering.
    • Experimental results demonstrate superior performance compared to standard k-means algorithms.
    • The algorithm effectively recovers clusters in both synthetic and real-world datasets.
    • The calculated variable weights can be directly applied to variable selection tasks.

    Conclusions:

    • The proposed algorithm enhances k-means clustering by incorporating automatic, data-driven variable weighting.
    • This approach offers a robust solution for clustering complex datasets and aids in variable selection for data mining.
    • The method shows significant improvements in cluster recovery accuracy over traditional algorithms.