Parallel coordinate order for <scp>high‐dimensional</scp> data

Visualization of high-dimensional data is counter-intuitive using conventional graphs. Parallel coordinates are proposed as an alternative to explore multivariate more effectively. However, it difficult extract relevant information through the parallel when with thousands overlapping lines. The order axes determines perception on coordinates. Thus, between attributes remains hidden if improperly ordered. Here we propose a general framework reorder This enough cover wide range visualization objectives. It also flexible contain many ordering measures. Consequently, present coordinate binary optimization problem and enhance achieve computationally efficient greedy approach that suits data. Our applied wine genetic purpose dimension reordering highlight attributes' dependence. Genetic reordered cluster detection. shows able adapt criteria for objective.