Welcome to the InfoSci Platform
Reference Hub10
An Enhanced Recursive Firefly Algorithm for Informative Gene Selection

An Enhanced Recursive Firefly Algorithm for Informative Gene Selection

Nassima Dif, Zakaria Elberrichi
Copyright: © 2019 |Volume: 10 |Issue: 2 |Pages: 13
ISSN: 1947-9263|EISSN: 1947-9271|EISBN13: 9781522566373|DOI: 10.4018/IJSIR.2019040102
Cite Article Cite Article

MLA

Dif, Nassima, and Zakaria Elberrichi. "An Enhanced Recursive Firefly Algorithm for Informative Gene Selection." IJSIR vol.10, no.2 2019: pp.21-33. http://doi.org/10.4018/IJSIR.2019040102

APA

Dif, N. & Elberrichi, Z. (2019). An Enhanced Recursive Firefly Algorithm for Informative Gene Selection. International Journal of Swarm Intelligence Research (IJSIR), 10(2), 21-33. http://doi.org/10.4018/IJSIR.2019040102

Chicago

Dif, Nassima, and Zakaria Elberrichi. "An Enhanced Recursive Firefly Algorithm for Informative Gene Selection," International Journal of Swarm Intelligence Research (IJSIR) 10, no.2: 21-33. http://doi.org/10.4018/IJSIR.2019040102

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Feature selection is the process of identifying good performing combinations of significant features among many possibilities. This preprocess improves the classification accuracy and facilitates the learning task. For this optimization problem, the authors have used a metaheuristics approach. Their main objective is to propose an enhanced version of the firefly algorithm as a wrapper approach by adding a recursive behavior to improve the search of the optimal solution. They applied SVM classifier to investigate the proposed method. For the authors experimentations, they have used the benchmark microarray datasets. The results show that the new enhanced recursive FA (RFA) outperforms the standard version with a reduction of dimensionality for all the datasets. As an example, for the leukemia microarray dataset, they have a perfect performance score of 100% with only 18 informative selected genes among the 7,129 of the original dataset. The RFA was competitive compared to other state-of-art approaches and achieved the best results for CNS, Ovarian cancer, MLL, prostate, Leukemia_4c, and lymphoma datasets.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.