MSBWO: A Multi-Strategies Improved Beluga Whale Optimization Algorithm for Feature Selection

Biomimetics, Journal Year: 2024, Volume and Issue: 9(9), P. 572 - 572

Published: Sept. 22, 2024

Feature selection (FS) is a classic and challenging optimization task in most machine learning data mining projects. Recently, researchers have attempted to develop more effective methods by using metaheuristic FS. To increase population diversity further improve the effectiveness of beluga whale (BWO) algorithm, this paper, we propose multi-strategies improved BWO (MSBWO), which incorporates circle mapping dynamic opposition-based (ICMDOBL) initialization as well elite pool (EP), step-adaptive Lévy flight spiral updating position (SLFSUP), golden sine algorithm (Gold-SA) strategies. Among them, ICMDOBL contributes increasing during search process reducing risk falling into local optima. The EP technique also enhances algorithm's ability escape from SLFSUP, distinguished original BWO, aims rigor accuracy development spaces. Gold-SA introduced quality solutions. hybrid performance MSBWO was evaluated comprehensively on IEEE CEC2005 test functions, including qualitative analysis comparisons with other conventional state-of-the-art (SOTA) approaches that were 2024. results demonstrate superior algorithms terms maintains better balance between exploration exploitation. Moreover, according proposed continuous MSBWO, binary variant (BMSBWO) optimizers obtained function ten UCI datasets random forest (RF) classifier. Consequently, BMSBWO has proven very competitive classification precision feature reduction.

Language: Английский

---

GQEO: Nearest Neighbor Graph-based Generalized Quadrilateral Element Oversampling for Class-imbalance Problem

Neural Networks, Journal Year: 2024, Volume and Issue: 184, P. 107107 - 107107

Published: Dec. 27, 2024

Language: Английский

---

FedITA: A cloud–edge collaboration framework for domain generalization-based federated fault diagnosis of machine-level industrial motors

Advanced Engineering Informatics, Journal Year: 2024, Volume and Issue: 62, P. 102853 - 102853

Published: Oct. 1, 2024

Language: Английский

---

Adaptive Model-Agnostic Meta-Learning Network for Cross-Machine Fault Diagnosis with Limited Samples

Published: Jan. 1, 2024

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

Language: Английский

---

A hybrid sampling algorithm for imbalanced and class-overlap data based on natural neighbors and density estimation

Knowledge and Information Systems, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 22, 2024

Language: Английский

---