Visualization in Software and Information Retrieval

Ricardo Baeza-Yates

Computer Science Department

Universidad de Chile

September, 1995

Design Techniques

• Organization:
  • Consistency
  • Use of the screen
  • Relations between objects
  • Navigation

• Economy (do the best with the less possible)
  • Simplicity (KISS)
  • Clarity
  • Distintiveness
  • Emphasis

• Communication:
  • Legibility
  • Easy to understand
  • Use of typography
  • Simbology
  • Use of color and texture

Basic Visualization Techniques

• Graphical abstractions and metaphors
• Multiple views (a must!)
• Spatial relations
  Example: Tree-Maps
• Fish-eye views
  Example: Graphs
• Hierarchical views

• Use of perspective (3D)
  Examples: Cone trees, authorship analysis
• Use of color or texture
  Examples: distinguish, highlight, relate views, order
• Use of sound
  Examples: visual enforcement, behavior, color replacing,
  exception conditions

Software Visualization

• Forces to characterize software
• Different levels and abstractions: systems, programs, algorithms
• Different types of elements: code, control, data, ...
• Several taxonomies and studies:
  • Price, Baecker and Small, Journal of Visual Languages, 1993

    Most complete

  • Stasko and Patterson, IEEE Visual Languages Conf, 1992

    Aspect, Abstractness, Animation and Automation

  • Roman and Cox, IEEE Software 1993

    Scope, abstraction, specification, interface and presentation

• Implementations are usually object-oriented

Visual Information Retrieval

• Visual Query Languages
  • R-links: query is taken from the text
  • Visual net
  • Using icons and two dimensional relations: or/and example
  • Using visualizations also as queries (see next)
• Visualizing one document
  • Self-similarity: Dotplot
  • Layout: Seesoft
• Visualizing one collection
  • Gravity metaphor: VIBE
  • Venn diagram metaphor: InfoCrystal
  • Spatial metaphor: Rooms
  • Graphs: associative networks
  • Two dimensional relations and color: Bookpile
• Visualizing queries