Subscribe
Issue No.03  March (2014 vol.20)
pp: 351364
Jenny Hyunjung Lee , Dept. of Comput. Sci., Stony Brook Univ., Stony Brook, NY, USA
Kevin T. McDonnell , Dept. of Math. & Comput. Sci, Dowling Coll., Oakdale, NY, USA
Alla Zelenyuk , Pacific Northwest Nat. Lab., Richland, WA, USA
Dan Imre , Imre Consulting, Richland, WA, USA
Klaus Mueller , Dept. of Comput. Sci., Stony Brook Univ., Stony Brook, NY, USA
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TVCG.2013.101
ABSTRACT
Although the euclidean distance does well in measuring data distances within highdimensional clusters, it does poorly when it comes to gauging intercluster distances. This significantly impacts the quality of global, lowdimensional space embedding procedures such as the popular multidimensional scaling (MDS) where one can often observe nonintuitive layouts. We were inspired by the perceptual processes evoked in the method of parallel coordinates which enables users to visually aggregate the data by the patterns the polylines exhibit across the dimension axes. We call the path of such a polyline its structure and suggest a metric that captures this structure directly in highdimensional space. This allows us to better gauge the distances of spatially distant data constellations and so achieve data aggregations in MDS plots that are more cognizant of existing highdimensional structure similarities. Our biscale framework distinguishes fardistances from neardistances. The coarser scale uses the structural similarity metric to separate data aggregates obtained by prior classification or clustering, while the finer scale employs the appropriate euclidean distance.
INDEX TERMS
Layout, Euclidean distance, Correlation, Indexes, Data visualization, Extraterrestrial measurements,visual analytics, Information visualization, multivariate visualization, clustering, highdimensional data
CITATION
Jenny Hyunjung Lee, Kevin T. McDonnell, Alla Zelenyuk, Dan Imre, Klaus Mueller, "A StructureBased Distance Metric for HighDimensional Space Exploration with Multidimensional Scaling", IEEE Transactions on Visualization & Computer Graphics, vol.20, no. 3, pp. 351364, March 2014, doi:10.1109/TVCG.2013.101
REFERENCES
