ParallAX - A data mining tool based on parallel coordinates

被引：11

作者：

Avidan, T ^{[1
]}

Avidan, S ^{[1
]}

机构：

[1] Tel Aviv Univ, Fac Exact Sci, IL-69978 Tel Aviv, Israel

来源：

COMPUTATIONAL STATISTICS | 1999年 / 14卷 / 01期

关键词：

visualization; parallel coordinates; data mining; multivariate data analysis;

D O I：

10.1007/PL00022707

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

ParallAX is a new user-friendly tool for visualizing and analyzing multivariate data using the parallel coordinates methodology. Complex queries are formed as logical combinations of a small number of atomic queries. Pre-processing facilities like banding, outlier-pruning, constrained permutations selection, wrapping and others are included. These and some more features are the basis towards automated search processes which are essential when categorizing data. In addition, the parallel coordinates display is dynamically linked to scatter plot displays, where the results of complex polygonal queries on one display are reflected on the other.

引用

页码：79 / 89

页数：11

共 7 条

[1] USE OF FACES TO REPRESENT POINTS IN K-DIMENSIONAL SPACE GRAPHICALLY [J].

CHERNOFF, H .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1973, 68 (342) :361-368

[2]

FAYYAD UM, 1996, ADV KNOWLEDGE DISCOV

[3]

INSELBERG A, 1990, PROCEEDINGS OF THE FIRST IEEE CONFERENCE ON VISUALIZATION - VISUALIZATION 90, P361, DOI 10.1109/VISUAL.1990.146402

[4]

MARTIN R, 1995, P IEEE C VIS ATL GA, P271

[5]

SCHMID C, 1994, P 7 SSDBM IEEE COMP

[6]

Ward M. O., 1994, Proceedings. Visualization '94 (Cat. No.94CH35707), P326, DOI 10.1109/VISUAL.1994.346302

[7]

WEGMAN E, 1990, J AM STAT ASS, V85

← 1 →