A process of metaheuristics aiming to explore an area of a search space of an optimization problem.
Published in Chapter:
Bio-Inspired Algorithms for Feature Selection: A Brief State of the Art
Rachid Kaleche (LIO Laboratory, Université Oran1, Algeria), Zakaria Bendaoud (GeCoDe Laboratory, University of Saida Dr. Moulay Tahar, Algeria), Karim Bouamrane (LIO Laboratory, Université Oran1, Algeria), and Khadidja Yachba (LIO Laboratory, Université Relizane, Algeria)
Copyright: © 2023
|Pages: 19
DOI: 10.4018/978-1-7998-9220-5.ch113
Abstract
Feature selection is an important process of machine learning, especially when facing the high dimensionality challenges due the unprecedent increase of data, namely big data. The main objective of the feature selection process is to find the smaller feature subset which optimizes a learning algorithm performance providing by this a better readability. The feature selection problem is known to be an NP-hard problem, and classical approaches tackling it reached their limits. Therefore, tackling feature selection problem by bio-inspired algorithms has gained an increased interest due to the improved obtained results. This study presents an overview of feature selection problem and bio-inspired algorithms as a background. Based on this background, bio-inspired algorithms modeling elements for feature selection are described, followed by application domains and bio-inspired algorithms approach samples for feature selection problems. In addition, challenges and issues are discussed aiming to open future research opportunities.