When using spatial data for decision-making, uncertainty will be prevalent - a good example is when using choropleth maps for visualisation. Regarding geographic data, uncertainty can be found in three categories: spatial, attribute and temporal. This seminar reports on research that explores the visual representation of spatial and attribute uncertainty in particular. Taking the choropleth map as the base representation, current visualisation of uncertainty techniques were added then surveyed (via the Internet) to ascertain how effective they were in this capacity. A new method was then introduced in a subsequent survey and compared. This visualisation technique utilises the output from hierarchical spatial data structures to express both spatial and attribute uncertainty simultaneously. Survey results are shown expressing the 'usability' and 'effectiveness' of this method. A more visually appealing variation is then proposed and surveyed (the Hexagonal or Rhombus [HoR] quadtree). The results also reveal how these novel visualisations can variously utilise a 'level of detail' or 'clutter' metaphor to effectively communicate uncertainty.
Last modified: Tuesday, 03-Aug-2004 11:44:27 NZST
This page is maintained by Caroline Wills.