Visualization and Data Mining for High Dimensional Datasets英文电子书.pdf

Chapter 14 VISUALIZATION AND DATA MINING FOR HIGH DIMENSIONAL DATASETS Alfred Inselberg Senior Fellow San Diego SuperComputing Center, California, USA Multidimensional GraphsLid Raanana, Ismel Departmenst of Computer Science and Applied Mathematics TelAviv University TelAviv, Israel aiisrealOmath.tau.ac.il Abstract Visualization provides insightthrough images and can be consideredas a collec- tion of applickon specificmappin ~roblern~omain-VisualRange. For the visualization of multivariate ~roblemsa multidimensional system of ~ a r - allel coordinates (abbr. 11-coords) is constructed which induces a one-to-one mapping between subsets of N-space and subsets of 2-space. The result is a rig- orous methodology for doing and seeingN-dimensional geometry. Startingwith an the overview of the mathematical foundations, it is seen that the display of high-dimensional datasets and search for multivariate relations among the vari- ables is transformed into a 2-D pattern recognition problem. This is the basis for the application to Visual Data Mining which is illustrated with a real dataset of VLSI (Very Large Scale Integration -chip) production. Then a recent geo- metric classifier is presented and applied to 3 real datasets. The results compared to those of 23 other classifiershave the least error. The algorithm, has quadratic computational complexity in the size and number of parameters, provides com- prehensible and explicit rules, does dimensionality selection -where the mini- mal set of original variables required to state the rule is found, and orders these variables so as to optimize the clarity of separation between the designated set and its complem