I. Introduction
Feature subset selection is a well-known pattern recognition problem which is usually viewed as a data mining enhancement technique. This technique aims to reduce the number of features to be used, i.e., to reduce the entire feature space to a highly predictive subset of the space. This reduction may improve the performance of data mining algorithms to be used, in terms of speed, accuracy, and simplicity. In addition, because of this reduction, the identification of features which do not need to be stored, collected or bought, may bring financial savings [19].