Filter/Wrapper Methods for Gene Selection and Classification of Microarray Dataset

Abstract

A wide variety of large-scale information has been made within the extraction of genomic information and the extraction of data. Problems addressed embody ordination sequencing, supermolecule structure modeling, or the reconstruction of biological process trees (phylogeny). These issues need collaboration between biologists and computer scientists as a result of the issues to be of nice recursive difficulties. One of the most modern problems that gene expression data is resolved is with feature selection. There are two general approaches for feature selection: filter approach and wrapper approach. In this article, the authors propose a new approach when combining the filter approach with method ranked information gain and a wrapper approach with the searching method of the genetic algorithm.in order to test their overall performance, an experimental study is presented based on two gene microarray datasets found in bioinformatics and biomedical domains leukemia, and the central nervous system (CNS). The classifier Decision tree (C4.5) is used for improving the classification performance. The results show that their approach selects genes for additional correct classification emphasizes the effectiveness of the chosen genes and its ability to filter the information from unsuitable genes.