The classification of tumor using gene expression profile based on support vector machines and factor analysis

Gene expression data that is being used to gather information from tissue samples is expected to significantly improve the development of efficient tumor diagnosis and to provide understanding and insight into tumor related cellular processes. In this paper we propose a novel feature selection approach which integrates the feature score criterion with factor analysis to further improve the SVM-based classification performance of gene expression data. We examine two sets of published gene expression data to validate the novel feature selection method by means of SVM classifier with different parameters. Experiments show that the proposed hybrid method can select a small quantity of principal factors to represent a large number of genes and SVM has a superior classification performance with the common factors which are extracted from gene expression data. Moreover experiment results demonstrate successful cross-validation accuracy of 92% for the colon dataset and 100% for the leukemia dataset.