Fractal Neural Network Feature Selector for Automatic Pattern Recognition System

Basabi CHAKRABORTY; Yasuji SAWADA

Fractal Neural Network Feature Selector for Automatic Pattern Recognition System

Basabi CHAKRABORTY, Yasuji SAWADA

Full Text Views

0

Cite this

Summary :

Feature selection is an integral part of any pattern recognition system. Removal of redundant features improves the efficiency of a classifier as well as cut down the cost of future feature extraction. Recently neural network classifiers have become extremely popular compared to their counterparts from statistical theory. Some works on the use of artificial neural network as a feature selector have already been reported. In this work a simple feature selection algorithm has been proposed in which a fractal neural network, a modified version of multilayer perceptron, has been used as a feature selector. Experiments have been done with IRIS and SONAR data set by simulation. Results suggest that the algorithm with the fractal network architecture works well for removal of redundant informations as tested by classification rate. The fractal neural network takes lesser training time than the conventional multilayer perceptron for its lower connectivity while its performance is comparable to the multilayer perceptron. The ease of hardware implementation is also an attractive point in designing feature selector with fractal neural network.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E82-A No.9 pp.1845-1850

Publication Date: 1999/09/25

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Section on Nonlinear Theory and Its Applications)

Category

Cite this

Copy

Basabi CHAKRABORTY, Yasuji SAWADA, "Fractal Neural Network Feature Selector for Automatic Pattern Recognition System" in IEICE TRANSACTIONS on Fundamentals, vol. E82-A, no. 9, pp. 1845-1850, September 1999, doi: .
Abstract: Feature selection is an integral part of any pattern recognition system. Removal of redundant features improves the efficiency of a classifier as well as cut down the cost of future feature extraction. Recently neural network classifiers have become extremely popular compared to their counterparts from statistical theory. Some works on the use of artificial neural network as a feature selector have already been reported. In this work a simple feature selection algorithm has been proposed in which a fractal neural network, a modified version of multilayer perceptron, has been used as a feature selector. Experiments have been done with IRIS and SONAR data set by simulation. Results suggest that the algorithm with the fractal network architecture works well for removal of redundant informations as tested by classification rate. The fractal neural network takes lesser training time than the conventional multilayer perceptron for its lower connectivity while its performance is comparable to the multilayer perceptron. The ease of hardware implementation is also an attractive point in designing feature selector with fractal neural network.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e82-a_9_1845/_p

Copy

@ARTICLE{e82-a_9_1845,
author={Basabi CHAKRABORTY, Yasuji SAWADA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Fractal Neural Network Feature Selector for Automatic Pattern Recognition System},
year={1999},
volume={E82-A},
number={9},
pages={1845-1850},
abstract={Feature selection is an integral part of any pattern recognition system. Removal of redundant features improves the efficiency of a classifier as well as cut down the cost of future feature extraction. Recently neural network classifiers have become extremely popular compared to their counterparts from statistical theory. Some works on the use of artificial neural network as a feature selector have already been reported. In this work a simple feature selection algorithm has been proposed in which a fractal neural network, a modified version of multilayer perceptron, has been used as a feature selector. Experiments have been done with IRIS and SONAR data set by simulation. Results suggest that the algorithm with the fractal network architecture works well for removal of redundant informations as tested by classification rate. The fractal neural network takes lesser training time than the conventional multilayer perceptron for its lower connectivity while its performance is comparable to the multilayer perceptron. The ease of hardware implementation is also an attractive point in designing feature selector with fractal neural network.},
keywords={},
doi={},
ISSN={},
month={September},}

Copy

TY - JOUR
TI - Fractal Neural Network Feature Selector for Automatic Pattern Recognition System
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1845
EP - 1850
AU - Basabi CHAKRABORTY
AU - Yasuji SAWADA
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E82-A
IS - 9
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - September 1999
AB - Feature selection is an integral part of any pattern recognition system. Removal of redundant features improves the efficiency of a classifier as well as cut down the cost of future feature extraction. Recently neural network classifiers have become extremely popular compared to their counterparts from statistical theory. Some works on the use of artificial neural network as a feature selector have already been reported. In this work a simple feature selection algorithm has been proposed in which a fractal neural network, a modified version of multilayer perceptron, has been used as a feature selector. Experiments have been done with IRIS and SONAR data set by simulation. Results suggest that the algorithm with the fractal network architecture works well for removal of redundant informations as tested by classification rate. The fractal neural network takes lesser training time than the conventional multilayer perceptron for its lower connectivity while its performance is comparable to the multilayer perceptron. The ease of hardware implementation is also an attractive point in designing feature selector with fractal neural network.
ER -