Open Access

ARTICLE

Microarray Gene Expression Classification: An Efficient Feature Selection Using Hybrid Swarm Intelligence Algorithm

Punam Gulande^*, R. N. Awale

Veermata Jijabai Technological Institute, Department of Electronics Engineering, Mumbai, 410210, India

* Corresponding Author: Punam Gulande. Email:

(This article belongs to the Special Issue: Impact of Internet of Medical Things (IoMT) on Smart and Secure Healthcare Applications)

Computer Systems Science and Engineering 2024, 48(4), 937-952. https://doi.org/10.32604/csse.2024.046123

Received 19 September 2023; Accepted 06 December 2023; Issue published 17 July 2024

Abstract

The study of gene expression has emerged as a vital tool for cancer diagnosis and prognosis, particularly with the advent of microarray technology that enables the measurement of thousands of genes in a single sample. While this wealth of data offers invaluable insights for disease management, the high dimensionality poses a challenge for multiclass classification. In this context, selecting relevant features becomes essential to enhance classification model performance. Swarm Intelligence algorithms have proven effective in addressing this challenge, owing to their ability to navigate intricate, non-linear feature-class relationships. This paper introduces a novel hybrid swarm algorithm, fusing the capabilities of the Artificial Bee Colony (ABC) and Firefly algorithms, aimed at improving feature selection in gene expression classification. The proposed method undergoes rigorous validation through statistical machine learning techniques and quantitative parameter evaluation, with comprehensive comparisons to established techniques in the field. The findings underscore the superiority of the hybrid Swarm Intelligence approach for feature selection in gene expression classification, offering promising prospects for enhancing cancer diagnosis and prognosis.

---

Keywords

---