Visual Analytics

Visual Analytics is an interdisciplinary approach that combines the advantages of different research areas. The goal of visual analytics method is to gain an understanding of extremely large and complex data sets. The approach combines the strengths of automatic data analysis with the capabilities of humans to comprehend namely fast patterns or trends visually. Through appropriate mechanisms of interaction data can be visually explored and knowledge is gained. It was introduced in 2004 and described a year later in the book " Illuminating the Path ".

Motivation

The ever growing amount of data to be processed has led to more and larger storage media have been developed. Often, the amount of data collected for later processing is however stored not filtered or cleaned up but as raw data. These data are in themselves useless, but can contain important information. With the help of visual analytics approach this flood of data is analyzed electronically, the person always has an influence on the automatically generated results. By means of suitable interactive visualizations man can direct the analysis process as desired. In contrast to pure information visualization the people will not only presents results but also have the opportunity he is given to intervene in the analysis and to influence the algorithms.

Process

Data: Heterogenous data sources have to be preprocessed before visual or automatic analysis first ( eg clean, normalize, etc.).

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

Visualization: be to verify the models by a user generated visualizations, which are enriched with interaction techniques for analysis.

The approach is geared to the following paradigm:

It is a continuous exchange between visual and automatic processes an important property of the Visual Analytics process. Adulterated results can be recognized early in order to get a better and more trustworthy result.

Areas of application

Application areas where large amounts have to be processed and visualized on data benefit from visual analytics.

These are for example:

Research institutions

  • Pacific Northwest National Laboratory ( PNNL )
  • National Center for Visual Analytics ( NCVA )
  • Chair for Databases, Data Analysis and Visualization, University of Konstanz
806697
de