Visual analytics

from Wikipedia, the free encyclopedia
The breadth of visual analytics

Visual analytics is an interdisciplinary approach that combines the advantages of different research areas. The aim of the visual analytics method is to gain knowledge from extremely large and complex data sets. The approach combines the strengths of automatic data analysis with the ability of humans to quickly grasp patterns or trends visually. Suitable interaction mechanisms allow data to be explored visually and insights gained. It was introduced in 2004 and described in the book "Illuminating the Path" a year later.

motivation

The visual analytics workflow. Based on DA Keim, J. Kohlhammer, GP Ellis, F. Mansmann: Mastering The Information Age - Solving Problems with Visual Analytics. Eurographics, 2010.

The steadily growing amount of data to be processed has led to ever larger storage media being developed. Often, however, the amount of data collected is neither filtered nor cleaned up for later processing, but instead is saved as raw data. This data is useless on its own, but it can contain important information. With the help of the visual analytics approach, this flood of data is analyzed electronically, whereby the human being always has an influence on the automatically generated results. Using suitable interactive visualizations , people can control the analysis process at will. In contrast to pure information visualization , people are not only presented with results, but are also given the opportunity to intervene in the analysis and influence the algorithms .

process

Data: Heterogeneous data sources must first be preprocessed (e.g. adjusted, normalized, etc.) before the visual or automatic analysis.

Models: With the help of data mining techniques, models of the original data are generated, which are then visualized for evaluation purposes or for further improvements.

Visualization: To check the models by a user, visualizations are generated, which are enriched with interaction techniques for an analysis.

The procedure is based on the following paradigm:

" Analysis First - Show the Important - Zoom, Filter and Analysis Further - Details on Demand "

A constant change between visual and automatic processes is an important characteristic of the visual analytics process. In this way, falsified results can be recognized at an early stage in order to obtain a better and more trustworthy end result.

application areas

Areas of application in which large amounts of data have to be processed and visualized benefit from visual analytics.

For example:

  1. Physics and Astronomy: Recognizing unexpected phenomena in huge and dynamic data streams.
  2. Disaster control: The analysis of an emergency situation in order to develop suitable countermeasures that help to limit the damage (natural disasters, etc.).
  3. Biology and Medicine: The analysis of large amounts of bio-data (the human genome etc.).
  4. Business intelligence : analysis of customer data.
  5. Policy Modeling and E-Government: Analysis of Data for Policy Making.

Research areas

The research on visual analytics examines numerous interdisciplinary aspects from data analysis to visual perception and human-computer interaction .

These are for example:

  1. Data mining and knowledge discovery in databases : analysis of heterogeneous data
  2. Information visualization: Computer based interactive visualization of abstract data
  3. Intelligent and Adaptive Systems : Systems that adapt to the knowledge and skills of the users
  4. Visual Perception : Research and results of human visual skills related to cognition , especially the cognitive tasks for exploration and analysis
  5. Usability and user experience : usability ( usability ) and user experience in terms of understanding and experiencing

Research institutions

Individual evidence

  1. ↑ DA Keim, F. Mansmann, J. Schneidewind, J. Thomas, H. Ziegler: Visual analytics: Scope and challenges. Visual Data Mining, 2008, pp. 76-90.
  2. ↑ D. Keim, S. North, C. Panse, M. Sips: Visual Data Mining in Large Geo-Spatial Point Sets. In: IEEE Computer Graphics and Application. No. 12, 2004, pp. 36-44.
  3. ↑ DA Keim, F. Mansmann, J. Schneidewind, J. Thomas, H. Ziegler: Visual analytics: Scope and challenges. Visual Data Mining, 2008, p. 82.
  4. ^ J. Kohlhammer, U. Proff, A. Wiener: Visual Business Analytics. Effective access to data and information. dpunkt.verlag, 2013.
  5. ^ Peter Sonntagbauer; Kawa Nazemi, Susanne Sonntagbauer, Giorgio Prister, Dirk Burkhardt (Eds.): Handbook of Research on Advanced ICT Integration for Governance and Policy Modeling. IGI Global, 2014. doi: 10.4018 / 978-1-4666-6236-0
  6. ↑ DA Keim, F. Mansmann, J. Schneidewind, J. Thomas, H. Ziegler: Visual analytics: Scope and challenges. Visual Data Mining, 2008, p. 88.
  7. ↑ DA Keim, F. Mansmann, J. Schneidewind, J. Thomas, H. Ziegler: Visual analytics: Scope and challenges. Visual Data Mining, 2008, pp. 76-77.
  8. ^ K. Nazemi: Adaptive Semantics Visualization. Eurographics Associations, 2014, pp. 15-78.
  9. ^ K. Nazemi: Adaptive Semantics Visualization. Eurographics Associations, 2014, pp. 30-105.
  10. ^ K. Nazemi: Adaptive Semantics Visualization. Eurographics Associations, 2014.
  11. ↑ Research Group Human-Computer Interaction and Visual Analytics (VIS), Darmstadt University of Applied Sciences. Retrieved April 30, 2019 (American English).

literature

  • JJ Thomas, KA Cook (Ed.): Illuminating the path: The research and development agenda for visual analytics. IEEE Computer Society 2005, ISBN 0-7695-2323-4 .

Web links