A Novel Feature Selection Algorithm Based on Aquila Optimizer for COVID-19 Classification - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

A Novel Feature Selection Algorithm Based on Aquila Optimizer for COVID-19 Classification

Résumé

To this day, the prevention of coronavirus disease is still an arduous battle. Medical imaging technology has played an important role in the fight against the epidemic. This paper is to perform feature selection on the CT image feature sets used for COVID-19 detection to improve the speed and accuracy of detection. In this work, the population-based intelligent optimization algorithm Aquila optimizer is used for feature selection. This feature selection method uses an S-shaped transfer function to process continuous values and convert them into binary form. And when the performance of the updated solution is not good, a new mutation strategy is proposed to enhance the convergence effect of the solution. Through the verification of two CT image sets, the experimental results show that the use of the S-shaped transfer function and the proposed mutation strategy can effectively improve the effect of feature selection. The prediction accuracy of the features selected by this method on the two open datasets is 99.67% and 99.28%, respectively.
Fichier sous embargo
Fichier sous embargo
0 5 14
Année Mois Jours
Avant la publication
mercredi 1 janvier 2025
Fichier sous embargo
mercredi 1 janvier 2025
Connectez-vous pour demander l'accès au fichier

Dates et versions

hal-04178744 , version 1 (08-08-2023)

Licence

Identifiants

Citer

Ling Li, Jeng-Shyang Pan, Zhongjie Zhuang, Shu-Chuan Chu. A Novel Feature Selection Algorithm Based on Aquila Optimizer for COVID-19 Classification. 12th International Conference on Intelligent Information Processing (IIP), May 2022, Qingdao, China. pp.30-41, ⟨10.1007/978-3-031-03948-5_3⟩. ⟨hal-04178744⟩
32 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More