Invention of the microarray technology has facilitated considerable improvement in the survival rate of the cancer patients. Microarray gene expression data has a small sample size and a large dimension. In this paper we suggest a hybrid combination of feature selection and feature extraction methods to reduce the size of gene expression data. Thresholding method is used for feature selection and discrete wavelet transform is used for feature extraction. The classification is performed using neural network algorithms. The results of classification are compared for different values of thresholds, wavelets and classification algorithms.
