Difference between revisions of "Post-editing"
Jump to navigation
Jump to search
(TODO: conduct experiment on post-editing speed for nob→nno software l10n) |
|||
Line 1: | Line 1: | ||
'''Post-editing''' means correcting the output of Machine Translation. |
'''Post-editing''' means correcting the output of Machine Translation. |
||
When using MT for creating text for publication (e.g. translating software or web sites, documentation, articles and so on), the MT output needs correction by someone who has good knowledge of the source language and native-speaker fluency in the target language. |
When using MT for creating text for publication (e.g. translating software or web sites, documentation, articles and so on), the MT output needs correction by someone who has good knowledge of the source language and native-speaker fluency in the target language.<ref>http://www.mt-archive.info/10/AMTA-2014-W2-Sanchez-Gijon.pdf actually investigates the effect of mother-tongue knowledge on post-editing quality.</ref> |
||
Post-editing of MT has been shown in various studies to decrease translation time and in some cases even significantly increase quality<ref>S. Green, J. Heer, and C.D. Manning, [http://www.spencegreen.com/pubs/green+heer+manning.chi13.pdf The efficacy of human post-editing for language translation]. In Proceedings of CHI. 2013, 439-448.</ref> compared to unassisted translation (a suitable [[Translation Memory]] may still be faster/better than MT). |
Post-editing of MT has been shown in various studies to decrease translation time and in some cases even significantly increase quality<ref>S. Green, J. Heer, and C.D. Manning, [http://www.spencegreen.com/pubs/green+heer+manning.chi13.pdf The efficacy of human post-editing for language translation]. In Proceedings of CHI. 2013, 439-448.</ref> compared to unassisted translation (a suitable [[Translation Memory]] may still be faster/better than MT). |
Revision as of 11:26, 9 February 2015
Post-editing means correcting the output of Machine Translation.
When using MT for creating text for publication (e.g. translating software or web sites, documentation, articles and so on), the MT output needs correction by someone who has good knowledge of the source language and native-speaker fluency in the target language.[1]
Post-editing of MT has been shown in various studies to decrease translation time and in some cases even significantly increase quality[2] compared to unassisted translation (a suitable Translation Memory may still be faster/better than MT).
Publications on post-editing with Apertium:
- François Masselot, Petra Ribiczey, & Gema Ramírez-Sánchez (2010) "Using the Apertium Spanish-Brazilian Portuguese machine translation system for localisation". EAMT 2010: Proceedings of the 14th Annual conference of the European Association for Machine Translation, 27-28 May 2010, Saint-Raphaël, France. Proceedings ed.Viggo Hansen and François Yvon; 8pp. [PDF, 577KB]; presentation: 23 slides [PDF, 569KB]
See also
Notes
- ↑ http://www.mt-archive.info/10/AMTA-2014-W2-Sanchez-Gijon.pdf actually investigates the effect of mother-tongue knowledge on post-editing quality.
- ↑ S. Green, J. Heer, and C.D. Manning, The efficacy of human post-editing for language translation. In Proceedings of CHI. 2013, 439-448.