Ranking human and machine summarization systems

  • Authors:
  • Peter Rankel;John M. Conroy;Eric V. Slud;Dianne P. O'Leary

  • Affiliations:
  • University of Maryland, College Park, Maryland;IDA/Center for Computing Sciences, Bowie, Maryland;University of Maryland, College Park, Maryland;University of Maryland, College Park, Maryland

  • Venue:
  • EMNLP '11 Proceedings of the Conference on Empirical Methods in Natural Language Processing
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Text Analysis Conference (TAC) ranks summarization systems by their average score over a collection of document sets. We investigate the statistical appropriateness of this score and propose an alternative that better distinguishes between human and machine evaluation systems.