[Corpora-List] Reports on progress in evaluation of translation quality

[Maxim Khalilov]

A series of recent publications analyze collaborative research
opportunities between the academic world and industry in the field of
automatic and human evaluation of translation quality.

What is the correlation between automatic metrics for translation
evaluation and the results of the human assessment? What do industrial
users think about BLEU and other automatic metrics? When BLEU should and
shouldn't be used for research purposes and in industrial settings?

These and many more questions relevant to everybody working on the
real-world machine translation projects are addressed in the article
written by Lucia Specia (University of Sheffield), along with two free and
one for-members-only reports. Find an overview of recent achievements in
the field of translation quality in academia and comparative analysis of
various evaluation methods at
http://translationautomation.com/reports/improving-automatic-quality-evaluation-metrics-through-industry-academia-collaboration

-- 
*
Maxim Khalilov, R&D*

maxim at tauslabs.com

*Skype: Maxim Khalilov (desperbcn)*

*Twitter: twitter.com/maximkhalilov*

*www.tausdata.org | www.translationautomation.com*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20130603/8a1ebf21/attachment.htm>
-------------- next part --------------
_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora