Cancer is one of the deadly diseases which accounts for high mortality. Early detection of cancer using the microarray technique has been evolving. Computational approaches for the prediction of cancer have recently advanced. Microarray data analysis is a tedious process due to its complex nature. Microarray data involves data with higher dimensionalities, fewer samples, presence of noise, and more features (genes). The above-stated issues lead to a decrease in classification accuracy. To enhance the accuracy, the convolutional neural network has been used for microarray classification. The proposed work classification involves the use of a convolutional neural network that achieved an accuracy of 89% with the ReLu activation function and also adds a dropout layer to enhance the classification performance. The proposed CNN-R&D approach has been performed well in the colon and leukemia microarray dataset.