J09-1003 significant recent work on the corpus-based evaluation for information ordering . In
E06-1045 suggests a further drawback to corpus-based evaluation : users may vary systematically
P04-1050 language generation . Previous corpus-based evaluations of the coherence of text according
E06-1040 Marciniak and Strube ( 2004 ) . Such corpus-based evaluations have sometimes been criticised
S15-2134 end-user task , as opposed to corpus-based evaluation where all temporal information
P13-3024 &#381;ilka for discussions . <title> A corpus-based evaluation method for Distributional Semantic
J97-1004 both subjected to quantitative , corpus-based evaluations . Kukich employed a corpus-based
P04-1052 domains , we were able to perform a corpus-based evaluation using the Penn WSJ Treebank (
E06-1040 cases prefer variation , whereas corpus-based evaluations give higher ratings to systems
J01-4003 psycholinguistic experiments . Our corpus-based evaluation showed that prepended phrases
J09-4008 in using automatically computed corpus-based evaluation metrics to evaluate Natural Language
P05-1071 ( 2004 ) perform a large-scale corpus-based evaluation of their approach . They use
J09-4008 practical consideration is that corpus-based evaluations require a corpus of human-written
J09-4008 Cahill and van Genabith 2006 ) . Corpus-based evaluation has been especially popular in
E06-1040 seems clear that for automatic corpus-based evaluation to work well , we need high-quality
J01-4000 Anaphor Resolution Algorithm A Corpus-Based Evaluation of Centering and Pronoun Resolution
J09-4008 the solution space . Automatic corpus-based evaluations are appealing in NLG , as in
P00-1051 Parameters of Centering Theory : a Corpus-Based Evaluation using Text from Application-Oriented
J98-3001 ( Yeh and Mellish 1997 ) ; by corpus-based evaluation ( Robin and McKeown 1996 ) ;
J01-4003 draft of this paper . <title> A Corpus-Based Evaluation of Centering and Pronoun Resolution
hide detail