Please use this identifier to cite or link to this item: http://localhost:8080/xmlui/handle/123456789/5201
Title: MULTI-OBJECTIVE METAHEURISTIC OPTIMIZATION ALGORITHMS FOR WRAPPER-BASED FEATURE SELECTION: A LITERATURE SURVEY (Article)
Authors: Gopalakrishnan, Anitha
Vadivel, Vinodhini
Keywords: Evolutionary computing
Feature seletion
Metaheuristic algorithms
Multi-objective optimization
Issue Date: Oct-2023
Publisher: Institute of Advanced Engineering and Science
Abstract: In the data mining and machine learning (ML) discipline, feature selection problem is considered among many researchers in the recent times. Feature selection process targets to minimize feature set number and maximize performance accuracy by identifying optimal features. Multiple objectives are considered while identifying the optimal feature hence multi-objective metaheuristic optimization algorithms (MOMOAs) are applied. In this study, literature review is performed MOMOAs-for solving wrapper feature selection problem (WFS). The literature review for solving WFS problem and discuss the challenges faced by the researchers in solving the feature selection problem. The literature review is performed on all relevant studies published in the last 12 years [2009-2022]. A detailed overview of the feature selection preliminaries, MOMOAs-WFS, role of the classifier in feature selection problem are presented. The outcome of this literature review is to highlight the existing works related to WFS problem using MOMOAs. Finally, the research areas for improvement are identified and emphasized for the scientists to survey in the field of MOMOAs.
URI: https://doi.org/10.11591/eei.v12i5.4757
ISSN: 20893191
Appears in Collections:b) 2023-Scopus Open Access (Pdf)



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.