Laplacian star coordinates for visualizing multidimensional data

Multidimensional data visualization is an interesting research field with many applications in ubiquitous all fields of sciences. Star coordinates are one of the most common information visualization techniques for visualizing multidimensional data. A star coordinate system is a linear transformation that maps a multidimensional data space into a two-dimensional visual space, unfortunately, involving a loss of information. In this paper, we proposed to improve standard star coordinates by developing the concept of Laplacian star coordinates for visualizing multidimensional data. The Laplacian star coordinate system is based on dimension axes placement according to their similarity, which improves the quality of data representation. We prove the efficiency and robustness of our methods by measuring the quality of the representations for several data sets.