Microarray experiments allow the recording of expression levels of genes simultaneously. A microarray is a glass slide on to which DNA molecules are fixed at specific locations in a ordered manner called features. In data classification problem, given a set of data and the corresponding class label, the objective is to retrieve the relation among the data of the same class. In this paper Mutual Information is used to select the informative genes and Support Vector Machine is used to classify the genes. Support Vector Machines are learning machines provided with a set of examples (inputs) and with the associated labels (outputs). The proposed MI based gene selection and SVM classification is applied to classify multi-class microarray datasets. The obtained results show that this method provides higher classification accuracy compared to existing methods.
Volume 11 | 07-Special Issue
Pages: 1169-1174