Publications

Integrating Spatial & Non-spatial Data in Visualization

H. Hauser

Abstract

New opportunities in data science, such as the consideration of cohort study data, require new approaches to the appropriate design of an effective visualization. We need to capitalize on successful solutions from previous research, of course, but we should also explore new strategies that challenge our already established mindset in visualization. In this talk, I address the specific challenge of integrating spatial and non-spatial data in visualization, in particular, when the spatial aspect of the data is of great importance to the user–-this could relate to the morphological information in a 3D medical scan or the geometrical aspects of flow features in a CFD simulation. In data visualizaiton, the actual mapping step–-from data to a visual form–-is certainly crucial and we should strive to optimally exploit the great opportunities that we have in designing this step. In data-intensive sciences, the study objects of interest are increasingly often represented by extensive and rich datasets (aka. big data)–-while traditionally the focus of visualization was on individual, static datasets, we now face dynamic data, representing entire ensembles of study entities, etc. Visualization gets a lot harder, when facing such new 'big data' challenges–-both on the designer sider as well as also on the user side. At the same time, however, also the potential for impact is increasing, which amounts to a fantastic motivation for new basic research in visualization.

H. Hauser, Integrating Spatial & Non-spatial Data in Visualization, 2015.
[BibTeX]

New opportunities in data science, such as the consideration of cohort study data, require new approaches to the appropriate design of an effective visualization. We need to capitalize on successful solutions from previous research, of course, but we should also explore new strategies that challenge our already established mindset in visualization. In this talk, I address the specific challenge of integrating spatial and non-spatial data in visualization, in particular, when the spatial aspect of the data is of great importance to the user–-this could relate to the morphological information in a 3D medical scan or the geometrical aspects of flow features in a CFD simulation. In data visualizaiton, the actual mapping step–-from data to a visual form–-is certainly crucial and we should strive to optimally exploit the great opportunities that we have in designing this step. In data-intensive sciences, the study objects of interest are increasingly often represented by extensive and rich datasets (aka. big data)–-while traditionally the focus of visualization was on individual, static datasets, we now face dynamic data, representing entire ensembles of study entities, etc. Visualization gets a lot harder, when facing such new 'big data' challenges–-both on the designer sider as well as also on the user side. At the same time, however, also the potential for impact is increasing, which amounts to a fantastic motivation for new basic research in visualization.
@MISC {Hauser2015Austria,
author = "Helwig Hauser",
title = "Integrating Spatial \& Non-spatial Data in Visualization",
howpublished = "Invited talk",
month = "October",
year = "2015",
abstract = "New opportunities in data science, such as the consideration of cohort study data, require new approaches to the appropriate design of an effective visualization. We need to capitalize on successful solutions from previous research, of course, but we should also explore new strategies that challenge our already established mindset in visualization. In this talk, I address the specific challenge of integrating spatial and non-spatial data in visualization, in particular, when the spatial aspect of the data is of great importance to the user---this could relate to the morphological information in a 3D medical scan or the geometrical aspects of flow features in a CFD simulation. In data visualizaiton, the actual mapping step---from data to a visual form---is certainly crucial and we should strive to optimally exploit the great opportunities that we have in designing this step. In data-intensive sciences, the study objects of interest are increasingly often represented by extensive and rich datasets (aka. big data)---while traditionally the focus of visualization was on individual, static datasets, we now face dynamic data, representing entire ensembles of study entities, etc. Visualization gets a lot harder, when facing such new 'big data' challenges---both on the designer sider as well as also on the user side. At the same time, however, also the potential for impact is increasing, which amounts to a fantastic motivation for new basic research in visualization.",
pdf = "pdfs/2015-10-14-HHauser-InvTalk.pdf",
images = "images/ThumbPicHHAustria2015.png",
thumbnails = "images/ThumbPicHHAustria2015.png",
location = "Vienna, Austria"
}
projectidprojectid

Media

Downloads

Full paper [PDF]