Pradeep Singh, A. Shukla, M. Vardhan
Nov 1, 2017
Citations
0
Influential Citations
6
Citations
Journal
2017 International Conference on Inventive Computing and Informatics (ICICI)
Abstract
The explosion of DNA microarray dataset in the scientific repository has been encouraging interdisciplinary research on ecology, computer science, and bioinformatics. The DNA microarray research field, prompt the development of a remarkable gene selection (GS) method for gene expression data classification, due to exponentially increase data size and attribute into gene expression data. Over the last decades, many gene selection technologies have been successfully used by the researchers in the classification of tumors or cancers. To choose the informative genes involved in different types of tumors remains critical challenges. In order to select the informative genes for better classification, hybrid gene selection method is recommended in this paper. It is a combination of Spearman's Correlation Coefficient (SCC) and Genetic Algorithm (GA) method called as SCCGA. For the diverse performance measurements, the experiments are conducted on three real-life datasets of a varied dimensionality and number of instances. The numerical experimental results demonstrate that the proposed SCCGA method significantly reduces the data features which give better performance in terms of classification accuracy.