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

SOM-based algorithms for qualitative variables.

Marie Cottrell1, Smaïl Ibbou, Patrick Letrémy

  • 1SAMOS-MATISSE UMR 8595, 90, rue de Tolbiac, F-75634 Paris Cedex 13, France. cottrell@univ-paris1.fr

Neural Networks : the Official Journal of the International Neural Network Society
|November 24, 2004
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

Batch and median neural gas.

Neural networks : the official journal of the International Neural Network Society·2006
Same author

Double quantization of the regressor space for long-term time series prediction: method and proof of stability.

Neural networks : the official journal of the International Neural Network Society·2004
Same author

Statistical tools to assess the reliability of self-organizing maps.

Neural networks : the official journal of the International Neural Network Society·2002
See all related articles

This study unifies methods for analyzing categorical survey data using Self-Organizing Maps (SOM). New approaches extend SOM

Area of Science:

  • Data Science
  • Machine Learning
  • Statistics

Background:

  • Self-Organizing Maps (SOM) are known for topology-preserving data clustering, extending Principal Component Analysis.
  • Traditional SOM algorithms are limited to processing real-valued data.
  • Categorical data, common in surveys, presents a challenge for standard SOM.

Purpose of the Study:

  • To present a unified framework for analyzing categorical data using SOM-based methods.
  • To extend SOM capabilities beyond real-valued data for survey analysis.
  • To introduce and consolidate methods for handling categorical variables within SOM.

Main Methods:

  • Kohonen Multiple Correspondence Analysis (KMCA): Analyzes categorical data based on modalities.
  • Kohonen Multiple Correspondence Analysis with individuals (KMCA_ind): Incorporates both individuals and modalities.

Related Experiment Videos

  • Kohonen algorithm on DISJunctive table (KDISJ): Integrates individuals and modalities using a disjunctive table approach.

    • Main Results:

    • The unified framework provides versatile tools for categorical data analysis.
    • KMCA, KMCA_ind, and KDISJ offer distinct yet complementary approaches to SOM for categorical data.
    • These methods enable topology-preserving clustering and analysis of complex survey datasets.

    Conclusions:

    • The presented methods effectively adapt SOM for categorical data analysis, overcoming previous limitations.
    • The unified approach enhances the applicability of SOM to diverse datasets, particularly surveys.
    • These techniques offer powerful alternatives for exploring relationships within categorical variables and individuals.