In recent years, gene expression data combined with machine learning methods revolutionized cancer classification which had been based solely on morphological appearance. However, the characteristics of gene expression data have very-high-dimensional and small-sample-size which lead to over-fitting of classification algorithms. We propose a novel gene expression classification model of multiple classifying algorithms with synthetic minority oversampling technique (SMOTE) using features extracted by deep convolutional neural network (DCNN). In our approach, the DCNN extracts latent features of gene expression data, then the SMOTE algorithm generates new data from the features of DCNN was implemented. These models are used in conjunction with classifiers that efficiently classify gene expression data. Numerical test results on fifty very-high-dimensional and small-sample-size gene expression datasets from the Kent Ridge Biomedical and Array Expression repositories illustrate that the proposed algorithm is more accurate than state-of-the-art classifying models and improve the accuracy of classifiers including non-linear support vector machines (SVM), linear SVM, k nearest neighbors and random forests.