A comparative study of neural network models

Authors:
S. Hudon;Y. Yan;W. Kinsner
Affiliations:
Department of Electrical Engineering University of Manitoba Winnipeg, Manitoba, Canada R3T 2N2 and Ministry of National Defense BAMEO/ASO, Canadian Force Base Winnipeg Westwin, Manitoba, Canada R3 ...;Department of Electrical Engineering University of Manitoba Winnipeg, Manitoba, Canada R3T 2N2;Department of Electrical Engineering University of Manitoba Winnipeg, Manitoba, Canada R3T 2N2
Venue:
Mathematical and Computer Modelling: An International Journal
Year:
1990

Citing 9
Cited 1

A massively parallel architecture for a self-organizing neural pattern recognition machine

Computer Vision, Graphics, and Image Processing
The capacity of the Hopfield associative memory

IEEE Transactions on Information Theory
Constructing an associative memory

BYTE
Back-propagation

BYTE
Bidirectional associative memories

IEEE Transactions on Systems, Man and Cybernetics
The ART of Adaptive Pattern Recognition by a Self-Organizing Neural Network

Computer
Neurocomputing: foundations of research

Neurocomputing: foundations of research
Neural Networks, II: What Are They and Why is Everybody So Interested in Them Now?

IEEE Expert: Intelligent Systems and Their Applications
A model of the carotid vascular system with stenosis at the carotid bifurcation

Mathematical and Computer Modelling: An International Journal

Forecasting box office revenue of movies with BP neural network

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.98

Visualization

Abstract

The recent proliferation of neural network models for pattern recognition and other applications has led to a need for benchmarking and a comparative study of such models in order to assist in the development of new models more pertinent to such specific applications. This paper presents a comparative study of the following three important types of neural network models: (i) the Bidirectional Associative Memory (BAM), (ii) the Back Propagation (BP), and (iii) the Adaptive Resonance Theory (ART). The models have been analyzed to establish their efficiency, accuracy, and adaptability. Simulation of a unified computer implementation of the models (the same language, the same computer, and the same benchmark) can reveal the essential differences in the performance of the models, rather than the differences due to their implementations.