PRISMA - A multidimensional information visualization tool

From InfoVis:Wiki
Jump to navigation Jump to search

Authors

  • Paulo Igor A.Godinho
  • Bianchi Serique Meiguins
  • Aruanda Simões Gonçalves Meiguins
  • Ricardo M. Casseb do Carmo
  • Marcelo de Brito Garcia
  • Leandro Hernandez Almeida
  • Rodrigo Lourenço

Short description

PRISMA is an information visualization tool based on multiple coordinated views to explore multidimensional datasets using the following techniques: treemap, scatterplot and parallel coordinates. PRISMA is extensible, portable and easy to maintain since it has been developed in Java using design patterns.
Paper Authors


Interface


This figure shows the main User Interface of Prisma after loading a dataset. All options are in Portoguese.

Main Functions

Treemap


Treemap: The user can set the attribute representing each rectangle size and color.

Scatterplot


The scatterplot technique is based on the cartesian coordinate system. The user can select which attributes are represented by the X or Y axis.

Parallel Coordinates


Each axis represent a dimension which is arranged parallel. The lines connect the attributes of the data items.

Coordination


PRISMA uses a single dataset for all views. The user may analyze data in each view individually or simultaneously in all views. Coordination between techniques is static, i.e., the coordination of each air of views is pre-defined. Color, for instance, is strongly coordinated since the same color is defined for all views. Size, on the other hand, is loosely-coordinated since it may be used in the scatterplot and treemap views but needs to be explicitly specified in each view. Other coordinated tasks are filter, selection, and details on demand.
Paper Authors


Selection


This figure shows the selection feature used on the starting dataset.


Details on demand


After selecting an attribute the user can gather additional information by using the details-on-demand function.

Filter


The filter dynamic interface. This function helps reducing the overload in the user interface if the dataset is very big (with the range filter in this example).

Other Functions:

PRISMA also includes:

  • Pie, Line and Bar graphics
  • a report generator
  • a export data function

Evaluation:

A usability test has been carried out with 11 persons (6 Computer Professionals and 5 General Users). Each person had to solve 4 tasks using PRISMA.


However, the user general evaluation of PRISMA was very good. At the end of the test, users were able to indicate many prominent applications in different professional and non-professional fields.
Paper Authors


References:

PRISMA - A Multidimensional Information Visualization Tool Using Multiple Coordinated Views