Using hybrid scatterplots for visualizing multi-dimensional data

Scatterplot visualization techniques are known as a useful method that shows the correlations of variables on the axes, as well as revealing patterns or abnormalities in the multidimensional data sets. They are often used in the early stage of the exploratory analysis. Scatterplot techniques have the drawback that they are not quite effective in showing a high number of dimensions where each plot in two-dimensional space can only present a pair-wise of two variables on the x-axis and y-axis. Scatterplot matrices and multiple scatterplots provide more plots that show more pair-wise variables, yet also compromise the space due to the space division for the plots. This chapter presents a comprehensive review of multi-dimensional visualization methods. We introduce a hybrid model to support multidimensional data visualization from which we present a hybrid scatterplots visualization to enable the greater capability of individual scatterplots in showing more information. Particularly, we integrate star plots with scatterplots for showing the selected attributes on each item for better comparison among and within individual items, while using scatterplots to show the correlation among the data items. We also demonstrate the effectiveness of this hybrid method through two case studies.