A density estimation approach to statistical pattern recognition is discussed. The pattern vector is split into two parts factoring a high dimensional class density function into a product of two lower dimensional density functions. The first factor, corresponding to the non-Gaussian structure in the data, is modeled nonparametrically. The second factor is modeled as a multivariate Gaussian conditionally on the first part of the pattern vector. Exploratory data analysis based on two-dimensional scatter plots is used to examine the plausibility of the density model. The proposed method is applied to the classification of handwritten digits and satellite image data.