Using Rough Sets with Heuristics for Feature Selection
Journal of Intelligent Information Systems
Rough Sets: Theoretical Aspects of Reasoning about Data
Rough Sets: Theoretical Aspects of Reasoning about Data
Reduction algorithms based on discernibility matrix: the ordered attributes method
Journal of Computer Science and Technology
Joint classifier and feature optimization for cancer diagnosis using gene expression data
RECOMB '03 Proceedings of the seventh annual international conference on Research in computational molecular biology
Learning rough set classifiers from gene expressions and clinical data
Fundamenta Informaticae
Gene discovery in leukemia revisited: a computational intelligence perspective
IEA/AIE'2004 Proceedings of the 17th international conference on Innovations in applied artificial intelligence
Reduct Generation and Classification of Gene Expression Data
ICHIT '06 Proceedings of the 2006 International Conference on Hybrid Information Technology - Volume 01
Accurate Cancer Classification Using Expressions of Very Few Genes
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Gene selection from microarray data for cancer classification-a machine learning approach
Computational Biology and Chemistry
Transactions on rough sets VII
Bioinformatics with soft computing
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Evolutionary Rough Feature Selection in Gene Expression Data
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Hi-index | 0.00 |
Indentification of informative gene subsets responsible for discerning between available samples of gene expression data is an important task in bioinformatics. Reducts, from rough sets theory, corresponding to a minimal set of essential genes for discerning samples, is an efficient tool for gene selection. Due to the compuational complexty of the existing reduct algoritms, feature ranking is usually used to narrow down gene space as the first step and top ranked genes are selected . In this paper, we define a novel certierion based on the expression level difference btween classes and contribution to classification of the gene for scoring genes and present a algorithm for generating all possible reduct from informative genes.The algorithm takes the whole attribute sets into account and find short reduct with a significant reduction in computational complexity. An exploration of this approach on benchmark gene expression data sets demonstrates that this approach is successful for selecting high discriminative genes and the classification accuracy is impressive.