Extending the METEOR Machine Translation Evaluation Metric to the Phrase Level

This paper presents METEOR-NEXT, an extended version of the METEOR metric designed to have high correlation with postediting measures of machine translation quality. We describe changes made to the metric's sentence aligner and scoring scheme as well as a method for tuning the metric's parameters to optimize correlation with humantargeted Translation Edit Rate (HTER). We then show that METEOR-NEXT improves correlation with HTER over baseline metrics, including earlier versions of METEOR, and approaches the correlation level of a state-of-theart metric, TER-plus (TERp).

• Type: Journal
• Year: 2010
• Where: NAACL
• Authors: Michael J. Denkowski, Alon Lavie
• Keywords: Computational Linguistics, Correlation Level, NAACL 2010, Sentence Aligner, Translation Edit Rate