|
Title:
|
EVALUATION OF PROJECTION TECHNIQUES USING HUBERTS Γ STATISTICS |
|
Author(s):
|
Dorina Marghescu |
|
ISBN:
|
978-972-8924-40-9 |
|
Editors:
|
Jörg Roth, Jairo Gutiérrez and Ajith P. Abraham (series editors: Piet Kommers, Pedro Isaías and Nian-Shing Chen) |
|
Year:
|
2007 |
|
Edition:
|
Single |
|
Keywords:
|
Projection techniques, visualization, evaluation, Huberts Γ statistic |
|
Type:
|
Short Paper |
|
First Page:
|
192 |
|
Last Page:
|
197 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Projection techniques reduce the data dimensionality by combining the original variables into a smaller number of new
dimensions, in a linear or nonlinear manner. The projection methods are particularly useful because they lend themselves
to visual representations of data, when the number of new dimensions is one, two or three. In this paper, the aim is to
evaluate different visualization techniques based on projection techniques with respect to their effectiveness in preserving
the inherent relationships and structure of the dataset. For this purpose, we investigate the use of the Huberts Γ statistics
for evaluating the fit between the distance matrices of original data and projected data. Moreover, we investigate the use
of the modified Huberts Γ statistics for evaluating the effectiveness of projection techniques in preserving the clustering
structure inherent in the dataset, if such structure is present. |
|
|
|
|
|
|
