ASIS&T 2006 START Conference Manager    

Spatialized Information Visualizations: a “Basstep” Approach to Application Design

Charles-Antoine Julien, France Bouthillier, John Leide

ASIS&T Annual Meeting - 2006 (ASIS&T 2006)
Austin, Texas, November 3-9, 2006


Abstract

Large result sets stemming from topical queries on the Web cause disorientation and must be reduced to a manageable size through exclusions (NOTing) and/or restrictions (ANDing) which also reject relevant documents. The information visualization (IV) technique known as spatialization is an automatic process creating a topographical map metaphor of the semantic information space where similar documents are presented as single groups of varying sizes placed on a surface so that distance between the groups reflects their pair-wise similarity. A spatialized Web corpus visualization provides overview and visual segmentation providing sharper relevance evaluation and disorientation mitigation. This communication reports on efforts to apply the BASSTEP approach to the design of IV tools for information retrieval (IR) by isolating and evaluating (using paper mock-ups) information features of a spatialized Web corpus in the context of Web IR. Six (6) semi-structured formative interviews provide a list of most expected initial user interpretations of four (4) features and three (3) additional interviews attempt to triangulate the findings when these features are integrated into a single interface mock-up. Our work shows that the application of the BASSTEP approach allows low cost user evaluations early in the design process improving the usability of visual IR tools.


  
START Conference Manager (V2.52.6)
Maintainer: rrgerber@softconf.com