Journal of Mathematical Sciences: Advances and ApplicationsAuthor's: Mi Zhou
Pages: [51] - [78]
Received Date: January 21, 2016
Submitted by:
DOI: http://dx.doi.org/10.18642/jmsaa_7100121627
Fisher score and genetic algorithm are widely used for feature
selection. However, some redundant features will be selected by Fisher
score, and the convergence properties may be worse if the initial
population of genetic algorithm is generated by a random manner. To
improve the performance of feature selection by Fisher score and
genetic algorithm, we propose a hybrid feature selection method, which
merging the advantages of Fisher score and genetic algorithm together.
It aims at utilizing the features\' Fisher score to generate the
initial population of genetic algorithm. To begin with, the Fisher
scores of all the features will be mapped into a specific interval by
a linear function, and then the rescaled Fisher scores will be
utilized to generate the initial population of genetic algorithm.
Finally, the initial population will be used in the subsequent
procedure of genetic algorithm to perform feature selection with
elitist strategy for reference. In this paper, we choose four data
sets of Sonar, WDBC, Arrhythmia, and Hepatitis to test the performance
of our algorithm. Feature subsets of the four data sets will be
selected by our algorithm, and then the dimensionality of data sets
will be reduced according to the selected feature subsets,
respectively. 1-NN classifier is used to classify the dimensionality
reduced data sets, and respectively, achieving the classification
accuracy of 
with ten-fold cross validation method. The
experiment results show that, compared to the performance of Fisher
score, genetic algorithm and Fisher score genetic algorithm, our
algorithm is fit for eliminate redundant features, and can select
discriminative features. Above all, our method is effective in feature
selection.
feature selection, Fisher score, genetic algorithm, elitist strategy.
