Year: 2006
Type: conference-paper
Abstract: As part of evaluating a summary automatically, it is usual to determine how much of the contents of one or more humanproduced 'ideal' summaries it contains.Previous automated methods such as ROUGE compare using fixed word ngrams, which are not ideal for a variety of reasons.In this paper we describe a framework in which summary evaluation measures can be instantiated and compared, and we implement a specific evaluation method using very small units of content, called Basic Elements, that address some of the shortcomings of ngrams.This method is tested on DUC 2003DUC , 2004DUC , and 2005 systems and produces very good correlations with human judgments.
Institutions: University of Southern California, Ritsumeikan University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W1530435204
