A New Hybrid and Ensemble Gene Selection Approach with an Enhanced Genetic Algorithm for Classification of Microarray Gene Expression Values on Leukemia Cancer

Leukemia cancer, like other types of cancer, is a deadly health condition that threatens the lives of many people around the world. Micro array data are used extensively to reveal the gene-cancer as well as gene–gene relationships of Leukemia cancer due to the fact that it allows the expression valu...

Full description

Saved in:

Bibliographic Details
Published in:	International journal of computational intelligence systems Vol. 13; no. 1; pp. 1554 - 1566
Main Authors:	Bilen, Mehmet, Işik, Ali H., Yiğit, Tuncay
Format:	Journal Article
Language:	English
Published:	Dordrecht Springer Netherlands 01-01-2020 Springer
Subjects:	Cancer classification Ensemble approach genetic algorithm Hybrid algorithm microarray leukemia gene selection Research Article Hybrid algorithm microarray leukemia gene selection Cancer classification Ensemble approach genetic algorithm
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Leukemia cancer, like other types of cancer, is a deadly health condition that threatens the lives of many people around the world. Micro array data are used extensively to reveal the gene-cancer as well as gene–gene relationships of Leukemia cancer due to the fact that it allows the expression value of thousands of genes to be revealed at once. However, the size of the high-dimensional data that the micro arrays accommodate makes it difficult to work with these data. In this study, a new approach was suggested in order to classify the micro arrays of leukemia cancer in a more efficient way by reducing the data size choosing the significant genes. This approach includes two steps: the ensemble step and the hybrid step. In the first step, a gene filtration process is carried out by creating an ensemble gene selection algorithm through Fisher correlation score, Wilcoxon rank sum, and information gain methods. In the second step, the feature selection phase step, the most successful genes among these genes are revealed by using an enhanced genetic algorithm. As a result of the classification process, the leave one out cross validation (LOOCV), 5-fold, and 10-fold cross validation results were found 100%, 98.57, and 97.14, respectively also 100% accuracy was obtained by 2 genes.
ISSN:	1875-6891 1875-6883 1875-6883
DOI:	10.2991/ijcis.d.200928.001