Support Vector Machine (SVM) is the most widely used classifier for performing the classification of a massive dataset. This research paper aims to improve the feature selection and classify the gene expression data by using the SVM classifier. And also aim to decrease the computational time of the SVM-RFE (support vector machine recursive feature elimination) algorithm by identifying more than one redundant genes and removing them in every iteration. Most of the gene expression profile contains an enormous number of features with few numbers of samples, to reduce the number of features before applying to the classifier for performing the classification; a feature selection algorithm is needed. The most effective algorithm used to perform the feature selection of microarray is, SVM-RFE. On every iteration, it generates the rank of the features and removes the very least ranked feature, which is the most irrelevant. Since the modified algorithm is used to remove more than one redundant features in every iteration. It will help to reduce the computational time and increase the accuracy prediction.
