In this talk, we introduce a new fast and efficient method to visually discover complex structures and data distribution in massive sized multivariate data set. The proposed method is based on two linear and nonlinear statistical measures that construct two informative projections of the data. According to the developed theories, the two projections capture significant information from the given data that mitigate the visualization and exploration of the hyper-dimensional massive data. Furthermore, pattern(s) of complex geometry such as linear, nonlinear, and even mixing of both linear and nonlinear structures can be well identified. Multiple outliers can be easily depicted. Our findings in the data can be accurately interpreted according to the theory of the plot. We demonstrate the efficiency of the method on several complex large sized simulated and real data sets.