A discourse-aware graph-based content-selection framework

  • Authors:
  • Seniz Demir;Sandra Carberry;Kathleen F. McCoy

  • Affiliations:
  • University of Delaware, Newark, DE;University of Delaware, Newark, DE;University of Delaware, Newark, DE

  • Venue:
  • INLG '10 Proceedings of the 6th International Natural Language Generation Conference
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents an easy-to-adapt, discourse-aware framework that can be utilized as the content selection component of a generation system whose goal is to deliver descriptive texts in several turns. Our framework involves a novel use of a graph-based ranking algorithm, to iteratively determine what content to convey to a given request while taking into account various considerations such as capturing a priori importance of information, conveying related information, avoiding redundancy, and incorporating the effects of discourse history. We illustrate and evaluate this framework in an accessibility system for sight-impaired individuals.