Visual Interactions with a Multidimensional Ranked List

Abstract

Performance analysis of an interactive visualization system generally requires an extensive user study, a method that is very expensive and that often yields inconclusive results. To do a successful user study, the researcher has to be well aware of the system's possibilities. In this paper we present a different kind of analysis. We show how the system behavior and performance could be investigated off line, without direct user intervention. We introduce an evaluation measure to assess the quality of a multidimensional visualization. Next, we suggest two methods for dealing with user's feedback. We also discuss the effect the dimensionality of the visualization has on the actual performance.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1998
Accession Number
ADA341114

Entities

People

  • Anton Leouski
  • James Allan

Organizations

  • University of Massachusetts Amherst

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Clustering
  • Computer Science
  • Data Science
  • Embedding
  • Feedback
  • Information Retrieval
  • Information Science
  • Materials
  • Mathematical Models
  • Models
  • Precision
  • Probability
  • Statistical Tests
  • Test And Evaluation
  • Three Dimensional
  • Two Dimensional
  • Visualizations

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Regression Analysis.
  • Systems Analysis and Design