臺灣博碩士論文加值系統

譯者處理大量翻譯文件時，機器翻譯會是有利的輔助工具。本研究行使兩種策略，前編輯策略 (pre-editing) 及後編輯策略 (post-editing)。然而，很少學者比較此兩種策略操作翻譯機器的利弊。本研究旨在比較使用語言控制策略 (CL strategy) 及後編輯策略 (PE strategy)，操作Google 機器翻譯時，其中的翻譯過程及差異；實驗文本為兩篇智慧財產權相關的文章。研究對象為五位不同翻譯研究所的研究生。本研究並採用有聲思考法 (Think-Aloud Protocol)，記錄受測者的翻譯過程；翻譯任務結束後，本研究者訪問受測者對此兩種策略的感想。研究結果顯示，只有一位受測者的兩次任務成果有顯著差異。受測者在語言控制策略中面臨的障礙包括翻譯單位、背景知識、原文理解的困難、及遵循語言控制規則發生的問題；在後編輯策略中面臨背景知識及用字選擇的障礙。Google 翻譯機器本身有所局限及缺失；然而，即使受測者沒有完全遵循語言控制規則，它仍然可以產出讀得通的句子，且文法結構正確。本研究結果提供使用兩種策略的利與弊，可供譯者及教授參考。

關鍵字：翻譯過程、語言控制策略、後編輯策略、技術文本

Machine translation engines are considered one mode of the promising auxiliary tools when translators deal with a great number of tasks. Two strategies are employed in this study, one is “pre-editing,” and the other is “post-editing.” However, few researchers have compared these two strategies in conjunction with MT. This experimental study aims at investigating the differences between Google Translate-aided human translation processes while applying controlled-language (CL) strategy and post-editing (PE) strategy respectively to translate two documents related to Intellectual Property Rights. The researcher invited five graduate students from different graduate institutes of translation and interpretation in Taiwan to conduct the tasks and adopts Think-Aloud Protocol to record their processes. After the tasks, the researcher interviewed them for their feedback. The results of evaluation showed significant differences in only one participant’s performance in CL task and PE task. The participants encountered barriers in CL task, including translation unit, lack of background knowledge, barrier of incomprehension, and other barriers of complying with CL rules, while barriers encountered in PE task include insufficient background knowledge and word choice. Although Google Translate has limitations and flaws, it still could generate readable and grammatical outcome even if the participants did not completely comply with CL rules. The results showed pros and cons of these two strategies which can serve as guidelines for practitioners in the translation industry and teachers who instruct machine translation courses.

Keywords：translation process, controlled-language strategy, post-editing strategy, technical text

TABLE OF CONTENTS
摘要 i
Abstract ii
Table of Contents iii
List of Tables vii
List of Figures ix
CHAPTER ONE INTRODUCTION 1
1.1 Background of the Study 3
1.2 Statement of the Problem 5
1.3 Purpose of the Study and Research Questions 6
1.4 Significance of the Study 8
CHAPTER TWO LITERATURE REVIEW 10
2.1 Application of Machine Translation 10
2.1.1 Google Translate 14
2.1.2 Alternative Real-time Machine Translation Systems 18
2.2 Human Translation vs. Machine Translation 20
2.2.1 Translation Unit 21
2.2.2 Accuracy of Translation 23
2.2.3 Cost and Productivity 24
2.2.4 Potential Drawbacks 26
2.3 Technical Text 29
2.3.1 Controlled Language 33
2.3.2 Post-editing 39
CHAPTER THREE METHODOLOGY 44
3.1 Participants 44
3.2 Instruments 45
3.2.1 Experimental Materials 46
3.2.2 Translation Tool: Google Translate 47
3.2.3 Two Strategies: Controlled-language Strategy and Post-editing Strategy 48
3.3 Data Collection 49
3.4 Data Analysis 51
CHAPTER FOUR RESULTS 56
4.1 Evaluation of Translation in Two Tasks 57
4.1.1 Translation Time Spent on CL Strategy and PE Strategy 61
4.2 Barriers and Solutions 62
4.2.1 Translation Unit in CL Task 63
4.2.2 Barriers of Background Knowledge in CL Task 70
4.2.3 Barriers of Incomprehension in CL Task 76
4.2.4 Other Barriers of Complying with CL Rules 83
4.2.5 Barriers of Background Knowledge and Solutions in PE Task 91
4.2.6 Word Choice 99
4.3 Benefits of Output of Google Translate 105
4.3.1 Benefits from CL Strategy 107
4.3.1.1 Suggestions on Employing CL Strategy 111
4.3.2 Benefits from PE Strategy 114
CHAPTER FIVE CONCLUSION 118
5.1 Implications 120
5.2 Limitations and Suggestions 121
REFERENCES 123
Appendix A Google Translate saved an Iranian’s life (TVBS News) 132
Appendix B Photos of mistranslation 134
Appendix C Experimental materials 135
Appendix D The exemplary translations for the two materials 137
Appendix E The participants’ outcome in CL task 139
Appendix F The participants’ outcome in PE task 145

English references
Arenas, A. G. (2012). Productivity and quality in the post-editing of outputs from translation memories and machine translation (Doctoral dissertation, Doctoral thesis. Tarragona: Universitat Rovira i Virgili).
Arnold, D., Balkan, L., Humphreys, R. L., Meijer, S., & Sadler, L.(1994) Machine Translation: An Introductory Guide, Oxford: NCC Blackwell, 1994.
Bernardini, S. (1999). Using think-aloud protocols to investigate the translation process: Methodological aspects. University of Bologna, Bologna.
Big Think (Producer). (2012, September 10). What’s Lost (And Found) In Machine Translation [Video clip]. Retrieved from https://www.youtube.com/watch?v=_Ic_gph-2hU
Cardey, S., Greenfield, P., & Wu, X. (2004). Designing a controlled language for the machine translation of medical protocols: The case of English to Chinese. In Machine Translation: From Real Users to Research (pp. 37-47). Springer Berlin Heidelberg.
Fuchs, N. E., & Schwitter, R. (1995). Specifying logic programs in controlled natural
language. arXiv preprint cmp-lg/9507009.
Garcia, I. (2011). Translating by post-editing: is it the way forward?. Machine
Translation, 25(3), 217-237.
Google (2010, July 9). Inside Google Translate [Video file]. Retrieved from https://www.youtube.com/watch?v=_GdSC1Z1Kzs
Google Translate Blog (2015, October 6). Translate text within apps thanks to the latest android update. [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/2015/10/translate-text-within-apps-thanks-to.html
Google Translate Blog (2015, August 20). Watch your language! 44 of them, actually. [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/2015/08/watch-your-language-44-of-them-actually.html
Google Translate Blog (2015, February 20). Celebrate Mother Language Day by joining the Google Translate Community [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/
Google Translate Blog (2015, January 14).Hallo, hola, olá to the new, more powerful Google Translate app [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/
Google Translate Blog (2014, December 11) Google Translate - 10 More Languages with your Help [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/
Google Translate Blog (2014, October 15). Translate web pages more easily with the new Translate Chrome Extension [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/
Google Translate Blog (2013, December 10). Google Translate - now in 80 languages [Web blog message]. Retrieved from http://googletranslate.blogspot.tw/2013/12/google-translate-now-in-80-languages.html
Google Translate Community Retrieved from https://docs.google.com/document/d/1dwS4CZzgZwmvoB9pAx4A6Yytmv7itk_XE968RMiqpMY/pub
Groves, M., & Mundt, K. (2015). Friend or foe? Google Translate in language for academic purposes. English for Specific Purposes, 37, 112-121.
Hutchins, W. J., & Somers, H. L. (1992). An introduction to machine translation(Vol. 362). London: Academic Press.
Hutchins W. J. (1995) Machine Translation: A Brief Histroy, Oxford: Pergamon Press, 1995. Pages 431-445
Hutchins W. J. and Somers H. L. (1997) An Introduction to Machine Translation, London, ACADEMIC PRESS LIMITED
Hutchins, J. (2001). Machine translation and human translation: in competition or in complementation. International Journal of Translation, 13(1-2), 5-20.
Kaji, H. (1999, September). Controlled languages for machine translation: State of the art. In Machine Translation Summit VII (pp. 37-39).
Kittredge, R. I. 2003. Sublanguages and controlled languages. In Ruslan Mitkov,
editor, The Oxford Handbook of Computational Linguistics, pages 430–447.
Koehn, P., & Germann, U. (2014, April). The impact of machine translation quality on human post-editing. In Workshop on Humans and Computer-assissted Translation (pp. 38-46).
Komeili, Z., & Rahimi, A. (2011). An investigation of the translation problems incurred by English-to-Persian machine translations:“Padideh, Pars, and Google Softwares”. Procedia-Social and Behavioral Sciences, 28, 1079-1082.
Koponen, M. (2012, June). Comparing human perceptions of post-editing effort with post-editing operations. In Proceedings of the Seventh Workshop on Statistical Machine Translation (pp. 181-190). Association for Computational Linguistics.
Koponen, M., Aziz, W., Ramos, L., & Specia, L. (2012). Post-editing time as a measure of cognitive effort. Proceedings of WPTP.
Kuhn, T. (2014). A survey and classification of controlled natural languages.
Computational Linguistics, 40(1), 121-170.
Künzli, A. (2006). Translation revision - A study of the performance of ten professional
translators revising a technical text. Linguistics Insights, Volume 46: Insights into
Specialized Translation, 193-212
Künzli, A. (2007): Translation revision: A study of the performance of ten professional
translators revising a legal text. In: Y. Gambier, M. Shlesinger & R. Stolze (eds.),
Doubts and Directions in Translation Studies. Selected Contributions from the EST Congress Lisbon 2004. Amsterdam (John Benjamins), 115-126.
Lee, J. S., & Liao, P. S. (2011). A comparative study of human translation and machine
translation with post-editing. 編譯論叢, 4(2), 105-149.
Microsoft (2014, July 21). Skype Translator demo from WPC 2014 [Video file]. Retrieved from http://research.microsoft.com/apps/video/default.aspx?id=225021&r=1
Microsoft. Microsoft Translator. Retrieved from http://www.microsoft.com/translator/products.aspx
Mymemory translated.net. [Official website]. Retrieved from http://mymemory.translated.net/doc/
Newmark, P. (1988). A textbook of translation. New York: Prentice-Hall International.
O’Brien, S., & Roturier, J. (2007). How portable are controlled language rules? A
comparison of two empirical MT studies. Proceedings of MT Summit XI, 345-352.
Och, F. J., & Ney, H. (2004). The alignment template approach to statistical machine
translation. Computational linguistics, 30(4), 417-449.
Plitt, M., & Masselot, F. (2010). A productivity test of statistical machine translation
post-editing in a typical localisation context. The Prague Bulletin of Mathematical
Linguistics, 93, 7-16.
Quah, C. K. (2006). Translation and technology, England.
Segev, E. (2010). Google and the Digital Divide: The Bias of Online Knowledge.
Elsevier. Oxford OX28 4BN, UK
Sheppard, F. (2011). Medical writing in English: the problem with Google Translate. La
Presse Médicale, 40(6), 565-566.
Shih (2002). Materials Development and Task Design for an MT Course. 翻譯學研究集刊, (7), 1-28.
Shih, C. L. (2007). Mapping out Students' Translation Process: An MT-specific
Comparative Study. 翻譯學研究集刊, (10), 163-190.
Shih, C. L. (2013). Adaptations in Controlled Cultural Writing for Effective Machine
Translation: A Register-specific Probe. Theory and Practice in Language
Studies, 3(7), 1093-1102.
Shih, C. L. (2013). Controlled Cultural Writing for Effective
Machine-translation-enabled Intercultural Communication on the Web. Cultus,
125.
Specia, L., Turchi, M., Cancedda, N., Dymetman, M., & Cristianini, N. (2009, May).
Estimating the sentence-level quality of machine translation systems. In13th
Conference of the European Association for Machine Translation (pp. 28-37).
van der Eijck, P. (1998). Controlled languages in technical documentation. InSelected
Papers from the Eight CLIN meeting (pp. 187-203).

中文文獻
中央新聞 (民104年5月14日)。復仇者聯盟2陸翻譯落漆 撐下去變趕緊跑。【中央新聞】Retrieved from http://appweb.cna.com.tw/WebM/Menu/acn/201505140074.aspx
史宗玲 (2004)。電腦輔助翻譯。 臺北: 書林。
史宗玲 (2013)。網頁書寫新文體─ 跨界交流 [快譯通]。台中：白象。
李子瑄 & 曹逢甫。(2009)。 漢語語言學。 正中出版發行。
思 果 (2009)。翻譯研究。台北：大地。
香港矽谷 (民103年9月1日)。 QTranslate 整合 Google Translate、Bing、Baido...線上翻譯服務的免費工具。【香港矽谷】Retrieved from
http://www.hksilicon.com/kb/articles/578422/QTranslateGoogle-TranslateBingBaido#.VAVkSmq1Nl4.facebook
張子銘(民104年5月18日)。伊朗男心肌梗塞 GOOGLE翻譯「波斯文」救命。【TVBS新聞網】。取自http://news.tvbs.com.tw/old-news.html?nid=580552
張簡守展 (2013)。Google Translate之功能、特性及其應用研究：以使用說明書英譯中為例 (未出版之碩士論文)。輔仁大學跨文化研究所。台北。
黃家琪 (2013)。控制性語言書寫及機器翻譯：以澎湖天后宮籤詩為例 (未出版之碩士論文)。國立高雄第一科技大學-口筆譯研究所。高雄。
葉子南 (2013)。英漢翻譯理論與實踐 第二版。台北：書林出版有限公司。
董大暉、藍月素、吳宜錚(2012)。譯後編輯用於提升機器翻譯品質之比較研究。
財團法人語言訓練測驗中心，pp. 317-338。
解志強 (2010)。 翻譯中的語言規範問題: 談網路的語料庫功能。編譯論叢， 3
(2)，163-202。
劉宓慶 (1997)。文體與翻譯。書林出版有限公司。
賴慈芸 (2008)。四種翻譯評量工具的比較。 編譯論叢，1(1)，71-91。
顏晴榮 (2007)。有聲思考法的應用。中等教育，58(4)，106-116。
騰訊科技/ 責任編輯:jimmonzang (民104年1月12日)。 翻譯要失業了! 科技巨頭
力翻譯技術【騰訊微博】Retrieved from
http://tech.qq.com/a/20150112/047731.htm