Pattern classification: a unified view of statistical and neural approaches
Pattern classification: a unified view of statistical and neural approaches
An introduction to variable and feature selection
The Journal of Machine Learning Research
Gene extraction for cancer diagnosis by support vector machines-An improvement
Artificial Intelligence in Medicine
A Survey on Filter Techniques for Feature Selection in Gene Expression Microarray Analysis
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Hi-index | 0.00 |
The paper develops the methods of selection of the most important gene sequence on the basis of the gene expression microarray, corresponding to different types of cancer. Special two stage strategy of selection has been proposed. In the first stage we apply few different methods of assessment of the importance of genes. Each method stresses different aspects of the problem. In the second stage the selected genes are compared and the genes chosen most frequently by all methods of selection are treated as the most important and representative for the particular type of problem. The results of selection are analyzed using PCA and the selected genes form the input to the SVM classifier recognizing the classes of cancer. The numerical experiments confirm the efficiency of the proposed approach.