Visual Analysis of North Atlantic Hurricane Trends using Parallel Coordinates and Statistical Techniques
Chad A. Steed, Patrick Fitzpatrick, T.J. Jankun-Kelly, and J. Edward Swan II. Visual Analysis of North Atlantic Hurricane Trends using Parallel Coordinates and Statistical Techniques. Memorandum Report #NRL/MR/7440--08-9130, Naval Research Laboratory, Stennis Space Center, Mississippi, 2008.
Download
Abstract
The integration of automated statistical analysis capabilities with a highly interactive, multivariate visualization interface is presented in this paper. Innovative visual interaction techniques such as dynamic axis scaling, conjunctive parallel coordinates, statistical indicators, and aerial perspective shading are exploited to enhance the utility of classical parallel coordinate plots. Moreover, the system facilitates statistical processes such as stepwise regression and correlation analysis to assist in the identification and quantification of the most significant predictors for a particular dependent variable. These capabilities are combined into a unique visualization system that is demonstrated via a North Atlantic hurricane climate study using a systematic workflow. This research corroborates the notion that enhanced parallel coordinates coupled with statistical analysis can be used for more effective knowledge discovery and confirmation in complex, real-world data sets.
BibTeX
@TechReport{TR08-htpcst,
author = {Chad A. Steed and Patrick Fitzpatrick and T.J. Jankun-Kelly
and J. Edward {Swan~II}},
title = {Visual Analysis of North Atlantic Hurricane Trends using
Parallel Coordinates and Statistical Techniques},
institution = {Naval Research Laboratory, Stennis Space Center, Mississippi},
type = {Memorandum Report},
number = {#NRL/MR/7440--08-9130},
date = {July 7},
month = {July},
year = 2008,
abstract = {
The integration of automated statistical analysis capabilities with a highly
interactive, multivariate visualization interface is presented in this
paper. Innovative visual interaction techniques such as dynamic axis scaling,
conjunctive parallel coordinates, statistical indicators, and aerial perspective
shading are exploited to enhance the utility of classical parallel coordinate
plots. Moreover, the system facilitates statistical processes such as stepwise
regression and correlation analysis to assist in the identification and
quantification of the most significant predictors for a particular dependent
variable. These capabilities are combined into a unique visualization system
that is demonstrated via a North Atlantic hurricane climate study using a
systematic workflow. This research corroborates the notion that enhanced
parallel coordinates coupled with statistical analysis can be used for more
effective knowledge discovery and confirmation in complex, real-world data sets.
},
}