The DNA microarray technology has modernized the approach of biology researchin such a way that scientists can now measure the expression levels ofthousands of genes simultaneously in a single experiment. Gene expressionprofiles, which represent the state of a cell at a molecular level, have greatpotential as a medical diagnosis tool. But compared to the number of genesinvolved, available training data sets generally have a fairly small samplesize for classification. These training data limitations constitute a challengeto certain classification methodologies. Feature selection techniques can beused to extract the marker genes which influence the classification accuracyeffectively by eliminating the un wanted noisy and redundant genes This paperpresents a review of feature selection techniques that have been employed inmicro array data based cancer classification and also the predominant role ofSVM for cancer classification.
展开▼
