Feature Selection Problem and Metaheuristics: A Systematic Literature Review about Its Formulation, Evaluation and Applications

Biomimetics, Journal Year: 2023, Volume and Issue: 9(1), P. 9 - 9

Published: Dec. 25, 2023

Feature selection is becoming a relevant problem within the field of machine learning. The feature focuses on small, necessary, and sufficient subset features that represent general set features, eliminating redundant irrelevant information. Given importance topic, in recent years there has been boom study problem, generating large number related investigations. this, this work analyzes 161 articles published between 2019 2023 (20 April 2023), emphasizing formulation performance measures, proposing classifications for objective functions evaluation metrics. Furthermore, an in-depth description analysis metaheuristics, benchmark datasets, practical real-world applications are presented. Finally, light advances, review paper provides future research opportunities.

Language: Английский

Multi-Strategy Improved Flamingo Search Algorithm for Global Optimization

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(9), P. 5612 - 5612

Published: May 1, 2023

To overcome the limitations of Flamingo Search Algorithm (FSA), such as a tendency to converge on local optima and improve solution accuracy, we present an improved algorithm known Multi-Strategy Improved (IFSA). The IFSA utilizes cube chaotic mapping strategy generate initial populations, which enhances quality set. Moreover, information feedback model is dynamically adjust based current fitness value, exchange between populations search capability itself. In addition, introduce Random Opposition Learning Elite Position Greedy Selection strategies constantly retain superior individuals while also reducing probability falling into optimum, thereby further enhancing convergence algorithm. We evaluate performance using 23 benchmark functions verify its optimization Wilcoxon rank-sum test. compared experiment results indicate that proposed can obtain higher accuracy better exploration abilities. It provides new for solving complex problems.

Language: Английский

Feature selection using modified chaotic satin bowerbird algorithm with deep transfer learning for Multispectral Image Classification

International Journal of Information Technology, Journal Year: 2024, Volume and Issue: unknown

Published: May 9, 2024

Language: Английский