In:Innovation and Expansion in Translation Process Research
Edited by Isabel Lacruz and Riitta Jääskeläinen
[American Translators Association Scholarly Monograph Series XVIII] 2018
► pp. 191–216
Chapter 9Human use of machine translation to extract information from texts
Published online: 1 February 2018
https://doi.org/10.1075/ata.18.09mic
https://doi.org/10.1075/ata.18.09mic
Abstract
As Machine Translation (MT) improves, technology is playing a larger role in translation work. Studies find productivity gains from using MT output, but questions remain about when MT is most helpful. Task-based evaluations are critical to understanding how users can leverage MT for tasks other than producing full translations. In the current experiment, English-dominant learners of Chinese extracted information from Chinese texts without MT, with static MT output, and with the ability to interact with the MT system. Although MT did not improve overall comprehension, participants accomplished the comprehension task faster with MT than without. Interacting with the MT system did not improve performance beyond using static MT output, suggesting that translators need training to realize maximum benefits from MT.
Article outline
- Introduction
- Challenges associated with the use of MT
- Anticipating MT errors
- Learning to interact with MT systems
- Challenges associated with the use of MT
- Overview of the current study
- Method
- Participants
- Design
- MT condition
- Type of task
- Task order
- Materials
- Source texts
- Relative quality of MT output across texts
- Participant ratings of texts
- Comprehension questions
- Procedure
- Results and discussion
- Participants’ language and translation experience
- Effects of MT condition on comprehension outcomes
- Comprehension outcome measures: Accuracy
- Comprehension outcome measures: Speed
- Statistical analysis approach
- Comprehension accuracy
- Comprehension speed
- Relationship between comprehension performance and MT use
- Relationship between comprehension performance and Chinese experience
- Effects of MT condition on translation outcomes
- Comparison of comprehension and translation results
- Use of resources
- General views of MT
- General discussion
- Conclusions
Notes References
References (20)
Bailey, Stacey, Carlos Balhana, Mohammad Khan, Keith Miller, and Adam Warner. 2013. “Automated and task-based evaluation of the effects of machine translation domain tuning on MT quality, post-editing, and human translation.” Technical report. McLean, VA: The MITRE Corporation.
Bates, Douglas, Martin Mächler, Ben Bolker, and Steve Walker. 2015. “Fitting linear mixed-effects models using lme4.” Journal of Statistical Software 67 (1): 1–48.
Callison-Burch, Chris. 2009. “Fast, cheap, and creative: Evaluating translation quality using Amazon’s Mechanical Turk.” In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, ed. by Philipp Koehn and Rada Mihalcea, 286–295. Stroudsburg, PA: Association for Computational Linguistics.
de la Fuente, Ruben. 2012. “Tips and tricks for full post editing.” Paper presented at the annual meeting of the American Translators Association, San Diego, California, October 24–27.
Drugan, Joanna. 2013. Quality in professional translation: Assessment and improvement. London: Bloomsbury.
Du, Xiaoyan. 2012. “A brief introduction of Skopos theory.” Theory and Practice in Language Studies 2: 2189–2193.
García, Ignacio. 2009. “Research on translation tools.” In Translation Research Projects 2, ed. by Anthony Pym and Alexander Perekrestenko, 27–33. Tarragona, Spain: Intercultural Studies Group.
Giannossa, Leonardo, Kirtee Vashee, Mike Dillinger, Elizabeth Lowe McCoy, and Patricia Phillips-Batoma. 2012. “Teaching Translation Studies students how MT works.” Paper presented at the annual meeting of the American Translators Association, San Diego, California, October 24–27.
Google Translate. 2014. Accessed October 1. [URL].
. 2015. Accessed February 20. [URL].
Jurafsky, Daniel, and James H. Martin. 2008. Speech and language processing, 2nd ed. Upper Saddle River, NJ: Prentice Hall.
Koponen, Maarit, Wilker Aziz, Luciana Ramos, and Lucia Specia. 2012. “Post-editing time as a measure of cognitive effort.” In Proceedings of the AMTA 2012 Workshop on Post-Editing Technology and Practice, ed. by Sharon O’Brien, Michel Simard, and Lucia Specia, 11–20. Association for Machine Translation in the Americas.
MacIntyre, Peter D., Kimberly A. Noels, and Richard Clément. 1997. “Biases in self-ratings of second language proficiency: The role of language anxiety.” Language Learning 47: 265–287.
Miller, Keith, Lucia Specia, Kim Harris, and Stacey Bailey, eds. 2014. Proceedings of Workshop on Automatic and Manual Metrics for Operational Translation Evaluation. [URL].
Ministry of Foreign Affairs of the People’s Republic of China. 2014. Accessed October 16. [URL].
O’Brien, Sharon. 2011. “Towards predicting post-editing productivity.” Machine Translation 25: 197–215.
Snover, Matthew, Nitin Madnani, Bonnie J. Dorr, and Richard Schwartz. 2009. “Fluency, adequacy, or HTER?: Exploring different human judgments with a tunable MT metric.” In Proceedings of the Fourth Workshop on Statistical Machine Translation, 259–268. Stroudsburg, PA: Association for Computational Linguistics.
Vela, Mihaela, Anne-Kathrin Schumann, and Andrea Wurm. 2014. “Human translation evaluation and its coverage by automatic scores.” In Proceedings of Workshop on Automatic and Manual Metrics for Operational Translation Evaluation, ed. by Keith Miller, Lucia Specia, Kim Harris, and Stacey Bailey, 20–30. [URL].