Pattern Recognition and Revealing using Parallel Coordinates Plot

Parallel coordinates plot (PCP) is an excellent tool for multivariate visualization and analysis, but it may fail to reveal inherent structures for datasets with a large number of items. In this paper, we propose a suite of novel clustering, dimension ordering and visualization techniques based on PCP, to reveal and highlight hidden structures. First, we propose a continuous spline based polycurves design to extract and classify different cluster aspects of the data. Then, we provide an efficient and optimal correlation based sorting technique to reorder coordinates, as a helpful visualization tool for data analysis. Various results generated by our framework visually represent much structure, trend and correlation information to guide the user, and improve the efficacy of analysis, especially for complex and noisy datasets.