Gene expression has a vast area of application in diagnosis of disease in the medical field. In this gene expression data, the number of genes involved is very large (in tenthousand) compared to the number of samples which is very few in cancer classification. This large number of genes in the training sample poses a challenge in cancer classification problem. An effective gene selection system is needed to choose a more relevant gene that plays an important role in cancer classification. Our research focuses on the efficient method of gene selection and cancer classification. An efficient gene selection method is needed to speed up the processing rate and increase the accuracy which in turn decreases the prediction rate. Particle Swarm Optimization (PSO) is used for selecting a subset of important genes which is used as an input for classification using improved Support-Vector Machine-Recursive FeatureElimination (SVM-RFE).
Kavitha K. R., U Harishankar, N., and Akhil, M. C., “PSO Based Feature Selection of Gene for Cancer Classification Using SVM-RFE”, in 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2018, pp. 1012-1016.