TY - GEN
T1 - Human effort and machine learnability in computer aided translation
AU - Green, Spence
AU - Wang, Sida
AU - Chuang, Jason
AU - Heer, Jeffrey
AU - Schuster, Sebastian
AU - Manning, Christopher D.
N1 - Publisher Copyright:
© 2014 Association for Computational Linguistics.
PY - 2014
Y1 - 2014
N2 - Analyses of computer aided translation typically focus on either frontend interfaces and human effort, or backend translation and machine learnability of corrections. However, this distinction is artificial in practice since the frontend and backend must work in concert. We present the first holistic, quantitative evaluation of these issues by contrasting two assistive modes: postediting and interactive machine translation (MT). We describe a new translator interface, extensive modifications to a phrasebased MT system, and a novel objective function for re-tuning to human corrections. Evaluation with professional bilingual translators shows that post-edit is faster than interactive at the cost of translation quality for French-English and English- German. However, re-tuning the MT system to interactive output leads to larger, statistically significant reductions in HTER versus re-tuning to post-edit. Analysis shows that tuning directly to HTER results in fine-grained corrections to subsequent machine output.
AB - Analyses of computer aided translation typically focus on either frontend interfaces and human effort, or backend translation and machine learnability of corrections. However, this distinction is artificial in practice since the frontend and backend must work in concert. We present the first holistic, quantitative evaluation of these issues by contrasting two assistive modes: postediting and interactive machine translation (MT). We describe a new translator interface, extensive modifications to a phrasebased MT system, and a novel objective function for re-tuning to human corrections. Evaluation with professional bilingual translators shows that post-edit is faster than interactive at the cost of translation quality for French-English and English- German. However, re-tuning the MT system to interactive output leads to larger, statistically significant reductions in HTER versus re-tuning to post-edit. Analysis shows that tuning directly to HTER results in fine-grained corrections to subsequent machine output.
UR - http://www.scopus.com/inward/record.url?scp=84961358859&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84961358859&partnerID=8YFLogxK
U2 - 10.3115/v1/d14-1130
DO - 10.3115/v1/d14-1130
M3 - Conference contribution
AN - SCOPUS:84961358859
T3 - EMNLP 2014 - 2014 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
SP - 1225
EP - 1236
BT - EMNLP 2014 - 2014 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
PB - Association for Computational Linguistics (ACL)
T2 - 2014 Conference on Empirical Methods in Natural Language Processing, EMNLP 2014
Y2 - 25 October 2014 through 29 October 2014
ER -