Unsupervised measurement of translation quality using multi-engine, bi-directional translation

Authors:
Menno van Zaanen;Simon Zwarts
Affiliations:
Division of Information and Communication Sciences, Department of Computing, Macquarie University, Sydney, NSW, Australia;Division of Information and Communication Sciences, Department of Computing, Macquarie University, Sydney, NSW, Australia
Venue:
AI'06 Proceedings of the 19th Australian joint conference on Artificial Intelligence: advances in Artificial Intelligence
Year:
2006

Citing 1
Cited 1

BLEU: a method for automatic evaluation of machine translation

ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics

Enhancing bilingual electronic group meeting comprehension with round-trip translations

International Journal of Information Systems and Change Management

Quantified Score

Hi-index	0.00

Visualization

Abstract

Lay people discussing machine translation systems often perform a round trip translation, that is translating a text into a foreign language and back, to measure the quality of the system. The idea behind this is that a good system will produce a round trip translation that is exactly (or perhaps very close to) the original text. However, people working with machine translation systems intuitively know that round trip translation is not a good evaluation method. In this article we will show empirically that round trip translation cannot be used as a measure of the quality of a machine translation system. Even when using translations of multiple machine translation systems into account, to reduce the impact of errors of a single system, round trip translation cannot be used to measure machine translation quality.