Dependency-based automatic evaluation for machine translation
Owczarzak, Karolina and van Genabith, Josef and Way, Andy (2007) Dependency-based automatic evaluation for machine translation. In: HLT-NAACL 2007 - Workshop on Syntax and Structure in Statistical Translation, 26 April 2007, Rochester, New York, USA.
Full text available as:
We present a novel method for evaluating the output of Machine Translation (MT), based on comparing the dependency
structures of the translation and reference rather than their surface string forms. Our method uses a treebank-based, wide coverage, probabilistic Lexical-Functional
Grammar (LFG) parser to produce a set of structural dependencies for each translation-reference sentence pair, and then calculates the precision and recall for these dependencies. Our dependency-based evaluation, in contrast to most popular string-based evaluation metrics, will not unfairly penalize perfectly valid syntactic variations in the translation. In addition to allowing for legitimate
syntactic differences, we use paraphrases in the evaluation process to account for lexical variation. In comparison with
other metrics on 16,800 sentences of Chinese-English newswire text, our method reaches high correlation with
human scores. An experiment with two translations of 4,000 sentences from Spanish-English Europarl shows that, in
contrast to most other metrics, our method does not display a high bias towards statistical models of translation.
Archive Staff Only: edit this record