The framework of the feature selection process. The feature selection methods are used in our analysis to choose the most related structural features of promoters. According to the relation with the target classifier, these methods are classified into two categories, filter and wrapper methods. Each method is composed of two critical elements: evaluation criteria of features and search strategy in the feature space.
Gan et al. BMC Bioinformatics 2012 13:4 doi:10.1186/1471-2105-13-4