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

Bicriterion methods for partitioning dissimilarity matrices.

Michael J Brusco1, J Dennis Cradit

  • 1Department of Marketing, College of Business, Florida State University, Tallahassee 32306-1110, USA. mbrusco@cob.fsu.edu

The British Journal of Mathematical and Statistical Psychology
|November 19, 2005
PubMed
Summary

This study introduces a bicriterion partitioning approach to address biases in cluster analysis. It balances cluster sums and partition diameter, offering a novel method for selecting optimal solutions in data partitioning.

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

A comparison of logistic regression methods for Ising model estimation.

Behavior research methods·2022
Same author

On maximization of the modularity index in network psychometrics.

Behavior research methods·2022
Same author

Disentangling relationships in symptom networks using matrix permutation methods.

Psychometrika·2021
Same author

On Fixed Marginal Distributions and Psychometric Network Models.

Multivariate behavioral research·2021
Same author

A comparison of 71 binary similarity coefficients: The effect of base rates.

PloS one·2021
Same author

Combinatorial Optimization of Clustering Decisions: An Approach to Refine Psychiatric Diagnoses.

Multivariate behavioral research·2020

Area of Science:

  • Data Science
  • Cluster Analysis
  • Optimization

Background:

  • Traditional partitioning indices can be biased by cluster size.
  • Minimizing partition diameter avoids size bias but may yield multiple, interpretively diverse optimal solutions.

Purpose of the Study:

  • To propose a bicriterion partitioning approach combining partition diameter and within-cluster sums.
  • To facilitate the selection of optimal solutions from alternative optima in data partitioning.

Main Methods:

  • Development of MATLAB-based exchange algorithms for bicriterion partitioning.
  • Evaluation of algorithms using synthetic datasets and an empirical dissimilarity matrix.

Main Results:

  • The bicriterion approach effectively balances partition diameter and within-cluster sums.

Related Experiment Videos

  • Developed algorithms rapidly provide excellent solutions for bicriterion partitioning problems.

    • Conclusions:

    • The proposed bicriterion approach offers a robust method for cluster analysis.
    • This facilitates more interpretable and reliable data partitioning by addressing limitations of existing methods.