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

Explicit magnification control of self-organizing maps for "forbidden" data.

Erzsébet Merényi1, Abha Jain, Thomas Villmann

  • 1Electrical and Computer Engineering Department, Rice University, Houston, TX 77005, USA. erzsebet@rice.edu

IEEE Transactions on Neural Networks
|May 29, 2007
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

Prototype-Based Classifiers and Vector Quantization on a Quantum Computer-Implementing Integer Arithmetic Oracles for Nearest Prototype Search.

Entropy (Basel, Switzerland)·2026
Same author

A fold switch regulates conformation of an alphavirus RNA-dependent RNA polymerase.

Nucleic acids research·2026
Same author

Minimal Polymerase-Containing Precursor Required for Chikungunya Virus RNA Synthesis.

Viruses·2025
Same author

Incorporation of arabinose-CTP and arabinose-UTP inhibits viral polymerases by inducing long pauses.

The Journal of biological chemistry·2025
Same author

Minimal polymerase-containing precursor required for Chikungunya virus RNA synthesis.

bioRxiv : the preprint server for biology·2025
Same author

[Development of a questionnaire to assess the effective factors of communicative movement therapy: IP-Kom62 and IP-Kom22].

Psychotherapie, Psychosomatik, medizinische Psychologie·2025

This study validates the Self-Organizing Map (SOM) magnification control for complex, high-dimensional data. Simulations show it effectively represents probability density functions and detects rare events, even when theory doesn't guarantee success.

Area of Science:

  • Data science
  • Machine learning
  • Artificial intelligence

Background:

  • The Self-Organizing Map (SOM) magnification control offers potential for data representation and rare event detection.
  • Theoretical limitations exist for its application to n-dimensional (n>=2) data with dependent components.

Purpose of the Study:

  • To assess the validity of the explicit SOM magnification control on theoretically unsupported data.
  • To investigate its performance on higher-dimensional and complex datasets.

Main Methods:

  • Systematic simulations were conducted on n-dimensional data where components are not statistically independent.
  • Performance was evaluated using synthetic and real-world datasets.

Main Results:

Related Experiment Videos

  • For unsupported 2D cases, the achieved magnification exponent systematically tracked the desired exponent with a positive offset.
  • Optimal probability density function (pdf) matching was achieved for simple synthetic higher-dimensional data.
  • Negative magnification improved the detectability of rare classes.

Conclusions:

  • The explicit SOM magnification control demonstrates practical utility beyond its theoretical guarantees.
  • It shows promise for analyzing complex, high-dimensional data, including real-world datasets.