AN ENHANCED FEATURE SELECTION METHOD BASED ON GREY WOLF OPTIMIZER FOR CLASSIFICATION PROBLEMS

This research emphasizes mainly on classification, in which every instance in the
dataset is classified into its target class depending on the information depicted by its
features. However, it is hard to select the suitable features from a set of features, because
the search space is generally large, wherein a dataset contains a number of features that
comprise redundant and unnecessary features, which leads in-turn to less performance
on the classification. Feature selection is considered the best way to solve this issue by
picking up only the most applicable features for the classification. In fact, feature
selection aims to remove redundant and irrelevant features and build the model more
efficiently. Feature selection categorized into two major types: wrapper and filter, this
thesis focuses only on wrapper approaches.