A computer implemented method is provided for clustering and visualization of an n-dimensional space in a Euclidean space. The method includes: collecting a plurality of multi-dimensional data sets; estimating a probability density function from each data set; approximating a dissimilarity between every pair of said probability density functions and recording every dissimilarity in a dissimilarity matrix; embedding the dissimilarity matrix into a Euclidean space having a dimensionality of three or less using a multi-dimensional scaling method; and graphically displaying relationships between data sets using data in the Euclidean space on a display of a computing device.