Autobrief: an experimental system for the automatic generation of briefings in integrated text and information graphics

  • Authors:
  • Nancy L. Green;Giuseppe Carenini;Stephan Kerpedjiev;Joe Mattis;Johanna D. Moore;Steven F. Roth

  • Affiliations:
  • Department of Mathematical Sciences, University of North Carolina of Greensboro, 383 Bryan Building, Greensboro, NC;Computer Science Department, University of British Columbia, Vancourer, BC, Canada CA;MAYA Viz. 2100 Wharton St., Suite 702, Pittsburgh, PA;MAYA Viz. 2100 Wharton St., Suite 702, Pittsburgh, PA;Human Communication Research Center, The University of Edinburgh, 2 Buccleuch Place, Edinburgh EH8 9LW, UK;MAYA Viz. 2100 Wharton St., Suite 702, Pittsburgh, PA and School of Computer Science, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA

  • Venue:
  • International Journal of Human-Computer Studies
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes AutoBrief. an experimental intelligent multimedia presentation system that generates presentations in text and information graphics in the domain of transportation scheduling. Acting as an intelligent assistant, AutoBrief creates a presentation to communicate its analysis of alternative schedules. In addition, the multimedia presentation facilitates data exploration through its complex information visualizations and support for direct manipulation of presentation elements. AutoBrief's research contributions include (1) a design enabling a new human-computer interaction style in which intelligent multimedia presentation objects (textual or graphic) can be used by the audience in direct manipulation operations for data exploration. (2) an application-independent approach to multimedia generation based on the representation of communicative goals suitable for both generation of text and of complex information graphics, and (3) an application-independent approach to intelligent graphic design based upon communicative goals. This retrospective overview paper, aimed at a multidisciplinary audience from the fields of human computer interaction and natural language generation, presents AutoBrief's design and design rationale.