A note on genetic algorithms for large-scale feature selection
Pattern Recognition Letters
Floating search methods in feature selection
Pattern Recognition Letters
Neural Networks for Pattern Recognition
Neural Networks for Pattern Recognition
Information Sciences: an International Journal - Recent advances in genetic fuzzy systems
Feature Selection for Knowledge Discovery and Data Mining
Feature Selection for Knowledge Discovery and Data Mining
Feature selection for high-dimensional genomic microarray data
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
CSB '03 Proceedings of the IEEE Computer Society Conference on Bioinformatics
Significance of Gene Ranking for Classification of Microarray Samples
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Data mining and genetic algorithm based gene/SNP selection
Artificial Intelligence in Medicine
Application of classification algorithms on IDDM rat data
ICDM'12 Proceedings of the 12th Industrial conference on Advances in Data Mining: applications and theoretical aspects
Hi-index | 0.00 |
Microarray gene expression data usually consist of a large amount of genes. Among these genes, only a small fraction is informative for performing cancer diagnostic tests. This paper focuses on effective identification of informative genes. A newly developed gene selection criterion using the concept of Bayesian discriminant is used. The criterion measures the classification ability of a feature set. Excellent gene selection results are then made possible. Apart from the cost function, this paper addresses the drawback of conventional sequential forward search (SFS) method. New genetic algorithms based Bayesian discriminant criterion is designed. The proposed strategies have been thoroughly evaluated on three kinds of cancer diagnoses based on the classification results of three typical classifiers which are a multilayer perception model (MLP), a support vector machine model (SVM), and a 3-nearest neighbor rule classifier (3-NN). The obtained results show that the proposed strategies can improve the performance of gene selection substantially. The experimental results also indicate that the proposed methods are very robust under all the investigated cases.