The Iterative Process of Interactive Visual Analysis
Helwig Hauser
MISC,
September, 2012
Abstract
One central characteristic of our information age is
that increasingly often we should exploit the wealth of available
data for the sake of learning, decision making, as well as other tasks.
A promising approach - not at the least also targeted by visual analytics -
is to integrate the strengths of computers (fast computation, efficient
handling of large datasets, comparably low costs, etc.) with the strengths
of the users (perceptual capabilities, considering domain knowledge,
detecting the unexpected, etc.). In this talk, we look at one possible
solution, i.e., the concept of interactive visual analysis, and describe
it as an iterative process, enabling the integration of computational and
interactive means for data exploration and analysis. We consider a data
scenario that opposes dependent and independent data dimensions (like in
a table), general enough to match many different application cases.
We focus on the case of multivariate data, but also address the case of
high-dimensional data and opportunities for exploring and analyzing such data.
After all, we think of interactive visual analysis as an iterative process,
where each step is performed on the basis of a toolbox with computational
and interactive visual solutions.
Published
Talk in the Visual Computing Forum (VCF) at UiB
Media
BibTeX
@misc{Hauser12VCF,
author = {Helwig Hauser},
title = {The Iterative Process of Interactive Visual Analysis},
year = {2012},
month = {September},
howpublished = {Talk in the Visual Computing Forum (VCF) at UiB},
location = {Bergen, Norway},
url ={http://www.ii.uib.no/vis/vcf/},
abstract = {One central characteristic of our information age is
that increasingly often we should exploit the wealth of available
data for the sake of learning, decision making, as well as other tasks.
A promising approach - not at the least also targeted by visual analytics -
is to integrate the strengths of computers (fast computation, efficient
handling of large datasets, comparably low costs, etc.) with the strengths
of the users (perceptual capabilities, considering domain knowledge,
detecting the unexpected, etc.). In this talk, we look at one possible
solution, i.e., the concept of interactive visual analysis, and describe
it as an iterative process, enabling the integration of computational and
interactive means for data exploration and analysis. We consider a data
scenario that opposes dependent and independent data dimensions (like in
a table), general enough to match many different application cases.
We focus on the case of multivariate data, but also address the case of
high-dimensional data and opportunities for exploring and analyzing such data.
After all, we think of interactive visual analysis as an iterative process,
where each step is performed on the basis of a toolbox with computational
and interactive visual solutions.},
}
www
