Multidimensional Data Visualization

Multidimensional Data Visualization

Dmitri Eidenzon (NovoSpark Corporation, Canada) and Olga Pilipczuk (University of Szczecin, Poland)
Copyright: © 2015 |Pages: 11
DOI: 10.4018/978-1-4666-5888-2.ch153
OnDemand PDF Download:
$30.00
List Price: $37.50

Chapter Preview

Top

Background

The history of the multidimensional visualization development is very long and can be divided into four stages:

  • 1.

    The beginning stage: From 1700 to 1970 - then the point symbols were used to show the geographical distribution of 56 commodities in Europe, scientists started to research on visual texture perception, and texture mapping works were created (Wong, & Bergeron, 1997);

  • 2.

    The forming stage: From 1970 to 1985 – the stage of new methods creation, such as Chernoff faces, parallel coordinates, iconography, words within words and many others; some of them were computerized (Wong, & Bergeron, 1997);

  • 3.

    The mining stage: From 1985 till now - modification and combination the well-known visualization methods with new elements; attempts to combine different types of expression, for example of sound with graphics, were made;

  • 4.

    The cognitive visualization stage: From 2000 till now – the dissemination of cognitive visualization conception and new cognitive multidimensional visualization methods creation.

Visualization methods can be divided into three groups (Wong, & Bergeron, 1997):

  • 1.

    Methods based on two-variate display: Include the fundamental two-variate displays and simultaneous views of two-variate displays. Most of them have been used in statistics. These techniques applied to data that have relatively small size, about hundreds of items.

  • 2.

    Methods based on multivariate display: Use colorful plots created by high-speed computer graphics computation. These techniques applied to data that are larger and more complicated than those used in techniques based on two-variate display.

  • 3.

    Animation methods: Are a multifunctional tool for visualizing multidimensional data. They include various movie animation models and techniques.

The traditional methods of multidimensional data visualization are: parallel coordinates, line graphs, survey plots, scatter plot matrix and it’s variations, star glyphs, treemaps, Sammon’s mapping, self-organizing map, dendrogram, radar chart, Voronoi diagrams, parallel glyphs, Bertin’s permutation matrices, Chernoff faces, worlds within worlds, table lens, VisDB, dynamic queries, attribute explorer et.

Key Terms in this Chapter

Cognitive Visualization: Multidimensional data visualization in a single image, which allow to find the source of the problem in a shorter period of time and contributed to the creation of new knowledge.

Dimensionality Reduction: Transforming the multidimensional data into a space of lower dimensions with preserving the relationships among them.

Multidimensional Data Visualization: Multidimensional data transformed into an image or a series of images.

Multidimensional Data: Data which related to more then two dimensions.

NovoSpark Curve: A two-dimensional image representing a multidimensional observation.

Cognitive Computer Graphics: Is a new science that integrates the cognitive science, computer graphics, psychology, graphic design, education science and many others depending on sphere of application.

NovoSpark Method: Is a new approach to the visualization of static and dynamic multidimensional observations.

Complete Chapter List

Search this Book:
Reset