A method for the visualization of environmental data via the analysis of correlations has been proposed. As an example, the data in correlation matrices of environmental parameters that describe air pollution in Vilnius city (10 parameters) as well as the development of coastal dunes and their vegetation in Finland (16 parameters) are visualized. These applied problems are very urgent because of their ecological nature: a visual presentation of data stored in the correlation matrices makes it possible for ecologists to discover additional knowledge hidden in the matrices and to make proper decisions. The method consists of two stages: building of a system of vectors based on the correlation matrix and its visualization. Sammon's mapping and the self-organizing map (two its realizations having different features for visualization) were applied in the visualization. The advantage of the method lies in the possibility to restore the system of multidimensional vectors describing variables from the correlation matrix—one vector for one variable. Copyright © 2004 John Wiley & Sons, Ltd.
[1]
Gintautas Dzemyda,et al.
Visualization of a set of parameters characterized by their correlation matrix
,
2001
.
[2]
Pirjo Hellemaa,et al.
The development of coastal dunes and their vegetation in Finland
,
1998
.
[3]
James C. Bezdek,et al.
An index of topological preservation for feature extraction
,
1995,
Pattern Recognit..
[4]
I. Jolliffe.
Principal Component Analysis
,
2002
.
[5]
John W. Sammon,et al.
A Nonlinear Mapping for Data Structure Analysis
,
1969,
IEEE Transactions on Computers.
[6]
A. Hörmann.
Software telegram: software package for multivariate analysis and multivariate time series analysis: Version 2.0.9 of Brodgar
,
2003
.
[7]
Jorma Laaksonen,et al.
SOM_PAK: The Self-Organizing Map Program Package
,
1996
.