Rasta (Rhetorical Structure Theory Analyzer), a discourse analysis component within the Microsoft English Grammar, efficiently computes representations of the structure of written discourse using information available in syntactic and logical form analyses. Rasta heuristically scores the rhetorical relations that it hypothesizes, using those scores to guide it in producing more plausible discourse representations before less plausible ones. The heuristic scores also provide a genre-independent method for evaluating competing discourse analyses: the best discourse analyses are those constructed from the strongest hypotheses.
%0 Report
%1 Corston-Oliver98beyond
%A Corston-Oliver, Simon H.
%D 1998
%K discourse, parsing, rst, structure
%T Beyond String Matching and Cue Phrases: Improving Efficiency and Coverage in Discourse Analysis
%U http://research.microsoft.com/pubs/view.aspx?tr\_id=204
%X Rasta (Rhetorical Structure Theory Analyzer), a discourse analysis component within the Microsoft English Grammar, efficiently computes representations of the structure of written discourse using information available in syntactic and logical form analyses. Rasta heuristically scores the rhetorical relations that it hypothesizes, using those scores to guide it in producing more plausible discourse representations before less plausible ones. The heuristic scores also provide a genre-independent method for evaluating competing discourse analyses: the best discourse analyses are those constructed from the strongest hypotheses.
@techreport{Corston-Oliver98beyond,
abstract = {{Rasta (Rhetorical Structure Theory Analyzer), a discourse analysis component within the Microsoft English Grammar, efficiently computes representations of the structure of written discourse using information available in syntactic and logical form analyses. Rasta heuristically scores the rhetorical relations that it hypothesizes, using those scores to guide it in producing more plausible discourse representations before less plausible ones. The heuristic scores also provide a genre-independent method for evaluating competing discourse analyses: the best discourse analyses are those constructed from the strongest hypotheses.}},
added-at = {2010-12-17T18:47:41.000+0100},
author = {Corston-Oliver, Simon H.},
biburl = {https://www.bibsonomy.org/bibtex/26cbc51afad3390313dc365a0f0449f7a/mortimer_m8},
citeulike-article-id = {238749},
citeulike-linkout-0 = {http://research.microsoft.com/pubs/view.aspx?tr\_id=204},
howpublished = {in papers from the AAAI Spring Symposium on Intelligent Text Summarizatio},
institution = {Microsoft Research},
interhash = {1a273cc16e7d2f4d4236f959a2e3389e},
intrahash = {6cbc51afad3390313dc365a0f0449f7a},
keywords = {discourse, parsing, rst, structure},
month = {November},
posted-at = {2005-06-27 14:50:23},
priority = {0},
timestamp = {2010-12-20T11:11:25.000+0100},
title = {{Beyond String Matching and Cue Phrases: Improving Efficiency and Coverage in Discourse Analysis}},
url = {http://research.microsoft.com/pubs/view.aspx?tr\_id=204},
year = 1998
}