Conversation clusters: grouping conversation topics through human-computer dialog

  • Authors:
  • Tony Bergstrom;Karrie Karahalios

  • Affiliations:
  • University of Illinois at Urbana-Champaign, Urbana, IL, USA;University of Illinois at Urbana-Champaign, Urbana, IL, USA

  • Venue:
  • Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.01

Visualization

Abstract

Conversation Clusters explores the use of visualization to highlight salient moments of live conversation while archiving a meeting. Cheaper storage and easy access to recording devices allows extensive archival. However, as the size of the archive grows, retrieving the desired moments becomes increasingly difficult. We approach this problem from a socio-technical perspective and utilize human intuition aided by computer memory. We present computationally detected topics of conversation as visual summaries of discussion and as reference points into the archive. To further bootstrap the system, humans can participate in a dialog with the visualization of the clustering process and shape the development of clustering models.