Visualization of high-dimensional data is counter-intuitive using conventional graphs. Parallel coordinates is proposed, as an alternative, to explore multivariate data more effectively. However, when the data are high-dimensional with thousands of lines overlapping, it is difficult to extract relevant information through the parallel coordinates. The order of the axes determines the perception of information on parallel coordinates. Thus, if coordinates are improperly ordered, the information between attributes remain hidden. Here we propose a general framework to reorder the coordinates. This framework depend on the objective of data visualization. It is also flexible to contain many conventional ordering measures. We also present the coordinate ordering binary optimization problem and enhance towards a computationally efficient greedy approach that suites high-dimensional data. Our approach is applied on wine data and on genetic data.
Published May 2017 , 14 pages