University of Bergen | Faculty of Mathematics and Natural Sciences | Department of Informatics | Visualization Group
Visualization
You are here: Department of Informatics > Visualization Group > Publications > Hauser12Dagstuhl
 Visualization
 > about
 > team & contact info
 > research
 > publications
 > projects
 > teaching
 > seminars
 > resources
 > network
 > events
 > links

Automated Methods in Information Visualization

Helwig Hauser

MISC, February, 2012

Abstract

Visualization and Machine Learning have related goals in terms of helping analysts to understand characteristic aspects of data. While visualization aims at involving the user through interactive depictions of data, machine learning is generally represented by automatic methods that yield optimal results with respect to certain initially specified tasks. Not at the least within the research direction of visual analytics it seems promising to think about opportunities to integrate both methodologies in order to exploit the strengths of both sides. Up to now, examples of integration very often encompass the visualization of results from automatic methods as well as attempts to make originally automated methods partially interactive. A vision for the future would be to integrate interactive and automatic methods in order to solve problems. A possible realization could be an iterative process where the one or other approach is chosen on demand at each step.

Published

Invited talk at the Dagstuhl seminar 12081

Media

  • presentation
  • www
  • Click to view

BibTeX

@misc{Hauser12Dagstuhl,
 author = {Helwig Hauser},
 title  ={Automated Methods in Information Visualization},
 year = {2012},
 month = {February},
 howpublished = {Invited talk at the Dagstuhl seminar 12081},
 location = {Wadern, Germany},
 url ={http://www.dagstuhl.de/en/program/calendar/semhp/?semnr=12081},
 abstract = {Visualization and Machine Learning have related goals in terms of helping
  analysts to understand characteristic aspects of data.  While visualization aims
  at involving the user through interactive depictions of data, machine learning
  is generally represented by automatic methods that yield optimal results with
  respect to certain initially specified tasks.  Not at the least within the
  research direction of visual analytics it seems promising to think about
  opportunities to integrate both methodologies in order to exploit the strengths
  of both sides.  Up to now, examples of integration very often encompass the
  visualization of results from automatic methods as well as attempts to make
  originally automated methods partially interactive.  A vision for the future
  would be to integrate interactive and automatic methods in order to solve
  problems.  A possible realization could be an iterative process where the one or
  other approach is chosen on demand at each step.   },

}






 Last Modified: Jean-Paul Balabanian, 2014-11-25