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 generic selection of features for neural network classifiers.

F Z Brill¹, D E Brown, W N Martin

¹Inst. for Parallel Comput., Virginia Univ., Charlottesville, VA.

IEEE Transactions on Neural Networks

|January 1, 1992

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

Cass County, (Mich.) Medical Society.

The North-Western medical and surgical journal·2023

Same author

Vaginal progesterone for prevention of preterm birth in asymptomatic high-risk women with a normal cervical length: a systematic review and meta-analysis.

The journal of maternal-fetal & neonatal medicine : the official journal of the European Association of Perinatal Medicine, the Federation of Asia and Oceania Perinatal Societies, the International Society of Perinatal Obstetricians·2021

Same author

Hypervitaminosis D and Metastatic Calcification in a Colony of Inbred Strain 13 Guinea Pigs, Cavia porcellus.

Veterinary pathology·2014

Same author

Mössbauer spectroscopy of europium-containing glasses: Optical activator study for x-ray image plates.

Hyperfine interactions·2014

Same author

Salmonella enterica causes more severe inflammatory disease in C57/BL6 Nramp1G169 mice than Sv129S6 mice.

Veterinary pathology·2013

Same author

Hot topic: association of telomere length with age, herd, and culling in lactating Holsteins.

Journal of dairy science·2012

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 efficient genetic algorithms for neural network feature selection. Novel techniques like approximate evaluation and training set sampling significantly reduce computation time for counterpropagation networks.

Area of Science:

Machine Learning
Artificial Intelligence
Computational Neuroscience

Background:

Neural network classifiers, particularly counterpropagation networks, require effective feature selection for optimal performance.
Traditional feature selection methods can be computationally intensive, limiting their practical application.

Purpose of the Study:

To develop and evaluate novel, computationally efficient genetic algorithm techniques for feature selection in counterpropagation networks.
To demonstrate the effectiveness of these optimized methods in identifying relevant features for neural network classification.

Main Methods:

Utilized a genetic algorithm framework for feature selection.
Implemented an approximate evaluation strategy using a nearest-neighbor classifier to assess feature sets.

Related Experiment Videos

Introduced a training set sampling method to accelerate the evaluation process.

Main Results:

The approximate evaluation using a nearest-neighbor classifier yielded features effective for counterpropagation networks.
Training set sampling enabled evaluations an order of magnitude faster than using the entire dataset.
Feature sets selected via training set sampling were comparable or superior to those selected using full dataset evaluation.

Conclusions:

Novel genetic algorithm approaches significantly enhance the efficiency of feature selection for counterpropagation networks.
Approximate evaluation and training set sampling offer substantial computational savings without compromising feature selection quality.
These optimized methods provide a practical solution for feature selection in complex neural network models.