As a commonly used technique in data preprocessing, feature selection selects a subset of informative attributes or variables to build models describing data. By removing redundant and irrelevant or noise features, feature selection can improve the predictive accuracy and the comprehensibility of the predictors or classifiers. Many feature selection algorithms with different selection criteria has been introduced by researchers. However, it is discovered that no single criterion is best for all applications. In this paper, we propose a framework based on a genetic algorithm (GA) for feature subset selection that combines various existing feature selection methods. The advantages of this approach include the ability to accommodate multiple feature selection criteria and find small subsets of features that perform well for a particular inductive learning algorithm of interest to build the classifier. We conducted experiments using three data sets and three existing feature selection methods. The experimental results demonstrate that our approach is a robust and effective approach to find subsets of features with higher classification accuracy and/or smaller size compared to each individual feature selection algorithm.
Abstract. Due to the huge number of genes and comparatively small number of samples from microarray gene expression data, accurate classification of diseases becomes challenging. Feature selection techniques can improve the classification accuracy by removing irrelevant and redundant genes. However, the performance of different feature selection algorithms based on different theoretic arguments varies even when they are applied to the same data set. In this paper, we propose a hybrid approach to combine useful outcomes from different feature selection methods through a genetic algorithm. The experimental results demonstrate that our approach can achieve better classification accuracy with a smaller gene subset than each individual feature selection algorithm does.
Microarray data usually contains a huge number of genes (features) and a comparatively small number of samples, which make accurate classification or prediction of diseases challenging. Feature selection techniques can help us identify important and irrelevant (unimportant) features by applying certain selection criteria. However, different feature selection algorithms based on various theoretical arguments often produce different results when applied to the same data set. This makes selecting an optimal or near optimal feature subset for a data set difficult. In this paper, we propose using a genetic algorithm to improve feature subset selection by combining valuable outcomes from multiple feature selection methods. The goal of our genetic algorithm is to achieve a balance between the classification accuracy and the size of the feature subsets selected. The advantages of this approach include the ability to accommodate different feature selection criteria and find small subsets of features that perform well for a particular inductive learning algorithm of interest to build the classifier. The experimental results demonstrate that our approach can find subsets of features with higher classification accuracy and/or smaller size compared with each individual feature selection algorithm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.