Evaluating Machine Translations Using mNCD

13 years 2 months ago

Download www.aclweb.org

This paper introduces mNCD, a method for automatic evaluation of machine translations. The measure is based on normalized compression distance (NCD), a general information theoretic measure of string similarity, and flexible word matching provided by stemming and synonyms. The mNCD measure outperforms NCD in system-level correlation to human judgments in English.

Marcus Dobrinkat, Tero Tapiovaara, Jaakko Väy

Real-time Traffic

ACL 2010 | Computational Linguistics | Information Theoretic Measure | MNCD Measure | System-level Correlation |

claim paper

» Unsupervised Measurement of Translation Quality Using Multiengine Bidirectional Translatio...

» Matura Evaluation Experiment Based on Human Evaluation of Machine Translation

» A Comparative Evaluation of Datadriven Models in Translation Selection of Machine Translat...

» Evaluation of ContextDependent Phrasal Translation Lexicons for Statistical Machine Transl...

» Machine Translation Method Using Inductive Learning with Genetic Algorithms

» A Dataset for Assessing Machine Translation Evaluation Metrics

» Feedback Cleaning of Machine Translation Rules Using Automatic Evaluation

» Using a Grammar Checker for Evaluation and Postprocessing of Statistical Machine Translati...

Post Info
More Details (n/a)

Added	10 Feb 2011
Updated	10 Feb 2011
Type	Journal
Year	2010
Where	ACL
Authors	Marcus Dobrinkat, Tero Tapiovaara, Jaakko Väyrynen, Kimmo Kettunen

Comments (0)

Sciweavers

Evaluating Machine Translations Using mNCD

ACL 2010 | Computational Linguistics | Information Theoretic Measure | MNCD Measure | System-level Correlation |

Explore & Download

Productivity Tools

Document Tools

Image Tools

Sciweavers