Input feature selection by mutual information based on Parzen window
N. Kwak, and C. Choi. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 24 (12):
1667--1671(2002)
Abstract
Mutual information is a good indicator of relevance between variables,
and have been used as a measure in several feature selection algorithms.
However, calculating the mutual information is difficult, and the
performance of a feature selection algorithm depends on the accuracy
of the mutual information. In this paper, we propose a new method
of calculating mutual information between input and class variables
based on the Parzen window, and we apply this to a feature selection
algorithm for classification problems.
%0 Journal Article
%1 Kwak2002
%A Kwak, N.
%A Choi, Chong-Ho
%D 2002
%J Pattern Analysis and Machine Intelligence, IEEE Transactions on
%K Parzen classification, density entropy, extraction, feature information information, mutual pattern probability selection, theory, window,
%N 12
%P 1667--1671
%T Input feature selection by mutual information based on Parzen window
%V 24
%X Mutual information is a good indicator of relevance between variables,
and have been used as a measure in several feature selection algorithms.
However, calculating the mutual information is difficult, and the
performance of a feature selection algorithm depends on the accuracy
of the mutual information. In this paper, we propose a new method
of calculating mutual information between input and class variables
based on the Parzen window, and we apply this to a feature selection
algorithm for classification problems.
@article{Kwak2002,
abstract = {Mutual information is a good indicator of relevance between variables,
and have been used as a measure in several feature selection algorithms.
However, calculating the mutual information is difficult, and the
performance of a feature selection algorithm depends on the accuracy
of the mutual information. In this paper, we propose a new method
of calculating mutual information between input and class variables
based on the Parzen window, and we apply this to a feature selection
algorithm for classification problems.},
added-at = {2009-09-12T19:19:34.000+0200},
author = {Kwak, N. and Choi, Chong-Ho},
biburl = {https://www.bibsonomy.org/bibtex/25922d999443877c4f1dcdf796de3bc77/mozaher},
file = {01114861.pdf:Kwak2002.pdf:PDF},
interhash = {a17c1ad54e43cfe8e5440b74e5ef4caa},
intrahash = {5922d999443877c4f1dcdf796de3bc77},
issn = {0162-8828},
journal = {Pattern Analysis and Machine Intelligence, IEEE Transactions on},
keywords = {Parzen classification, density entropy, extraction, feature information information, mutual pattern probability selection, theory, window,},
number = 12,
owner = {mozaher},
pages = {1667--1671},
timestamp = {2009-09-12T19:19:40.000+0200},
title = {Input feature selection by mutual information based on Parzen window},
volume = 24,
year = 2002
}