In the contemporary phase of big data, data visualization is one of the challenging segment of the discovery process. For a classifier, the primary goal is to identify the hidden levels of big data. The performance of classifiers depends on the feature space, number of classes and size of the data. To improve the reliability, efficiency and accuracy of the classifiers, new algorithms are required for analysis. This paper enables classification and visualization of information on diseases using a Deep Learning based Convolution Neural Network classifier. For feature selection and handling massive data in analysis of multivariate data is performed using particle swarm optimization (PSO) and principal component analysis (PCA) techniques. Real-world datasets are utilized for demonstration of the proposed learning algorithm. Deep learning classifiers are scientifically higher and offers better performance when compared to other classifiers according to the comparative study. © Published under licence by IOP Publishing Ltd.