Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Fast self-organizing feature map algorithm.

M C Su¹, H T Chang

¹Department of Electrical Engineering, Tamkang University, Tamkang, Taiwan, R.O.C. muchun@ee.tku.edu.tw

IEEE Transactions on Neural Networks

|February 6, 2008

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

Dual heterogeneous structures lead to ultrahigh strength and uniform ductility in a Co-Cr-Ni medium-entropy alloy.

Nature communications·2020

Same author

First trimester placental vascular indices and volume by three-dimensional ultrasound in pre-gravid overweight women.

Placenta·2019

Same author

Cinnamon effectively inhibits the activity of leukemia stem cells.

Genetics and molecular research : GMR·2016

Same author

A duplex SYBR Green I real-time quantitative PCR assay for detecting Escherichia coli O157:H7.

Genetics and molecular research : GMR·2013

Same author

Effect of phenethyl isothiocyanate on Ca2+ movement and viability in MDCK canine renal tubular cells.

Human & experimental toxicology·2012

Same author

Mathematical modeling of biofilm on activated carbon.

Environmental science & technology·2011

Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013

Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011

Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011

Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011

Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011

Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011

See all related articles

This study introduces an efficient three-stage method for creating feature maps quickly. It uses K-means clustering and a heuristic strategy, followed by Kohonen self-organizing map (SOM) fine-tuning for rapid, topologically ordered map generation.

Area of Science:

Machine Learning
Artificial Intelligence
Data Science

Background:

Traditional Kohonen self-organizing feature map (SOM) algorithms require extensive iterations for weight fine-tuning.
Feature map formation is crucial for data visualization and analysis.

Purpose of the Study:

To present an efficient, rapid approach for forming topologically ordered feature maps.
To reduce the computational cost and time associated with conventional SOM algorithms.

Main Methods:

A three-stage process involving K-means clustering for initial center selection.
A heuristic assignment strategy to create an initial feature map from selected data points.
Fine-tuning using the Kohonen SOM algorithm with a fast cooling regime.

Related Experiment Videos

Main Results:

The proposed method significantly accelerates the formation of feature maps compared to traditional SOM.
Topologically ordered feature maps are generated efficiently, reflecting data density distributions.
Validation performed on three distinct data sets demonstrates the method's effectiveness.

Conclusions:

The presented three-stage approach offers a computationally efficient alternative for feature map generation.
This method overcomes the limitations of conventional SOM by reducing iteration requirements.
The rapid formation of topologically ordered maps has implications for various data analysis applications.