A method of dual-process sample selection for feature selection on gene expression data
Authors:
- Liu QuanjinLQLiu Quanjin
- Zhao ZhiminZZZhao Zhimin
- Li Ying XinLXLi Ying Xin
- Yu XiaoleiYXYu Xiaolei
Open Access
- 9 May 2013
- journal article
- Published by Academic Journals in International Journal of Physical Sciences
- Vol. 8 (17) , 809-816
Abstract
A method of dual-process sample selection based on support vector machine (SVM) is proposed to select informative features in this paper. Samples in a training set are used to train a SVM model, and the samples excluding support vector samples are chosen to select critical features in the procedure of recursive feature elimination (RFE). The effect of the dual-process sample selection method on feature selection is evaluated using the classification and the clustering performance of the selected features. The proposed dual-process sample selection method is applied to five gene expression datasets, and the experimental results show that the method is useful to improve the performance of the feature selection method based on fuzzy interactive self-organizing data algorithm (ISODATA). This indicates the method is reliable and effective for selecting informative genes from gene expression data. Key words: Feature selection, support vector machine, fuzzy interactive self-organizing data algorithm (ISODATA), dual-process sample selection.Keywords
This publication has 14 references indexed in Scilit:
- Robust biomarker identification for cancer diagnosis with ensemble feature selection methodsBioinformatics, 2009
- SVMs Modeling for Highly Imbalanced ClassificationIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2008
- Recursive Fuzzy Granulation for Gene Subsets Extraction and Cancer ClassificationIEEE Transactions on Information Technology in Biomedicine, 2008
- Applying Support Vector Machines to Imbalanced Datasets2004
- Feature selection based on a modified fuzzy C-means algorithm with supervisionInformation Sciences, 2003
- Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learningNature Medicine, 2002
- Gene Selection for Cancer Classification using Support Vector MachinesMachine Learning, 2002
- Sample selection via clustering to construct support vector-like classifiersIEEE Transactions on Neural Networks, 1999
- Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by oligonucleotide arraysProceedings of the National Academy of Sciences, 1999
- Pattern Recognition with Fuzzy Objective Function Algorithms1981