Approche Filtre pour la sélection des gènes pertinents des données biopuces du Cancer du Côlon
Résumé
Developments in biotechnology have enabled biological measure of the information contained in thousands of genes using the DNA chip. This has identified the genes expressed in a given condition. The volume and specificity of these data sets consist of more features then the number of samples and lead to treatments involving the techniques of dimension reduction (features selection) to determine relevant and informative genes. In this work we focus on the selection of genes for colon cancer, that poses a major public health problem worldwide and especially in Algeria. We offer different types of features selection methods of statistical and probabilistic nature. Our experiments have shown that the approach has the ability to select a reduced number of features while preserving a very satisfactory classification rates.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...