Communication science et technologie
Volume 23, Numéro 2, Pages 36-43
2025-11-18
A Comparative Study Of Bio-inspired Algorithms For Feature Selection In Biomedical Data
Authors : Khellat Kihel Badra . Chouraqui Samira .
Abstract
Feature selection is a critical step in biomedical data analysis due to the high dimensionality and noise commonly present in datasets such as medical images and patient records. This paper presents a comparison of several bio-inspired optimization algorithms: Genetic Algorithm (GA), Artificial Immune System (AIS), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Artificial Bee Colony (ABC), Firefly Algorithm (FFA), Grey Wolf Optimizer (GWO), and Black Widow Optimization (BWO) for feature selection in biomedical classification tasks. These algorithms are evaluated on their ability to reduce feature dimensionality while maintaining or enhancing predictive performance. Experiments are conducted on benchmark biomedical datasets using k nearest neighbour (Knn) classifier. Our results highlight trade-offs between exploration and exploitation in different algorithms, revealing that FFA and GWO achieve higher accuracy with fewer features compared to swarm-based methods like PSO and ABC.
Keywords
Metaheuristics ; Feature Selection ; Disease diagnosis ; Medical Data ; Bio-inspired Algorithms
Les articles similaires
Btissem Gasmi
.
Hassina Seridi-bouchelaghem
.
Fouzia Anguel
.
pages 36-40.
Abderrahmane Gaham
.
Ahlem Melouah
.
Mohamed Amine Yakoubi
.
pages 95-102.
Benhammou Aissa
.
Hartani Mohamed Amine
.
Kimi Imade Eddine
.
Soumeur Mohammed Amine
.
Baba Mohammed
.
pages 26-32.
Amrane Ali
.
Zennir Youcef
.
pages 14-19.
Arbia W
.
Arbia L
.
Adour L
.
Lounici H
.
Mameri N
.
pages 29-36.
