臺灣博碩士論文加值系統

本研究旨在探討使用前機器翻譯編輯(前MT)文言文籤詩的有效策略。為進行研究，於前機器翻譯編輯的過程中，我使用控制性語言（CL）來編輯文言文籤詩，控制性語言之特性乃是限制其詞彙範圍和句子結構之多樣性。本研究以台灣籤詩為例，並利用語料庫/統計方法運作的Google Translate (線上Google翻譯) 當作翻譯工具，且用它來測試機器翻譯的品質。該項研究發現一些策略可運用在編輯文言文籤詩的過程，以提昇其機器譯文之品質。這些策略包括：(1)明朗化策略，例如：增添詞彙、添加轉折詞、改寫四字成語且添加典故；(2)常態化策略，例如：增加介系詞、添加冠詞及量詞，和(3)簡單化策略，例如：使用中文雙音節詞編譯、添加逗號或分號將一個句子分成兩個子句、改寫古典詞彙和四字片語為一般敘述。於前機器翻譯編輯後，籤詩之機器翻譯品質的確大幅地提升。經一些主修口筆譯的研究生、在何嘉仁美語的外籍老師們及一位英語教授閱讀機器譯文後，發現其文法正確性、語義清晰度和文化詞彙合宜性，皆比先前未經過編輯之機器譯文進步許多。最重要的是本研究發現控制性語言的使用有助於提高機器翻譯的品質且符合Gutt的關聯理論。關聯理論的重點是藉由改寫來源語特定的語言特性和特定的文化詞彙，使得來源語文本讓目標語讀者能在譯文中尋求與其文本認知與文化背景之關聯性，進而了解文本之訊息意涵。最後，本研究可應用在未來其它文類之有效機器翻譯相關研究。

This research aims to investigate the strategies for editing oracle poems from classical Chinese into vernacular controlled Chinese to improve its performance of machine translation (MT). The controlled language (CL), restricted in the scope of vocabulary and sentence structure, is used to edit oracle poems and Google Translate, a corpus/statistics-based MT tool, is used to test the quality of the MT output. The finding indicates that some strategies can be used to edit oracle poems, including: 1) explicitation strategies such as adding words, adding a transitional, and rewriting the Chengyu; 2) normalization strategies such as adding a preposition, and adding an article or/and a quantifier, and 3) simplification strategies such as using dual-syllable Chinese words, adding comma or semicolon to divide one sentence into two clauses, rewriting classical diction, and rewriting four-character phrases into common expressions.
The controlled poems after pre-MT editing did highly improve the grammatical accuracy, semantic clarity and pragmatic appropriateness of MT outputs based on the reading feedbacks from some students who major in translation and interpreting, some native English speakers teaching at HESS cram school and one English professor. Above all, this research takes on some significance, suggesting that the use of controlled language for better MT performance supports Gutt’s relevance theory that translation must seek contextual and cultural relevance to the target audience by modifying the SL-specific linguistic features and source-culture-specific idiomatic expressions. Due to the effectiveness of pre-MT editing with a host of strategies, this research model can be replicated in the future researches on editing other genres of texts for effective MT application.

摘要……………………………………………………………………….i
Abstract………………………………………………………………….ii
Table of Contents……………………………………………………….vi
Acknowledgements……………………………………………………..v
Chapter 1 Introduction…………………………………………………1
1.1 Research Motivation………………………………………………………....2
1.2 Research Purpose and Questions……..……………………………………..5
1.3 Research Method............................................................................................6
1.4 Research Structure……………………………………………………………9
Chapter 2 Literature Review………………………………………...10
2.1 Machine Translation………………………………………………………….10
2.1.1 The History of Machine Translation………………………………...11
2.1.2 Rule-based Machine Translation…………………………………….11
2.1.3 Corpus-based Machine Translation………………………………….12
2.2 Controlled Language and Pre-MT Editing……………………………….15
2.2.1 Controlled Language………………………………………………...15
2.2.2 Pre-MT Editing………………………………………………………18
2.3 Gutt’s Relevance Theory……………………………………………………19
Chapter 3 Lexical Strategies for Pre-editing Oracle Poems………...23
3.1 The Explicitation Strategies………………………………………………...23
3.1.1 Use of Dual-Syllable Words…………………………………………23
3.1.2 Adding Words………………………………………………………..27
3.2 The Normalization Strategy………………………………………………...34
3.2.1 Adding Word「是」as a Be Verb…………………...……………….34
3.2.2 Adding a Quantifier………………………………………………….37
3.2.3 Adding a Preposition………………………………………………...40
3.3 Implications of Applying Pre-MT Editing Strategies at the Lexical Level…44
Chapter 4 Syntactic Strategies for Pre-editing Oracle Poems………46
4.1 The Simplification Strategy…………………………………………………46
4.2 The Normalization Strategy……………………………….………………..50
4.3 The Explicitation Strategy…………………………………………………..55
4.4 Implications of Applying Pre-MT Editing Strategies at the Syntactic Level.60
Chapter 5 Strategies for Pre-editing Cultural-specific Items in Oracle Poems…………………………………………………………………...62
5.1 The Explicitation Strategies……………………………………….………..63
5.1.1 Rewriting Chengyu and Adding Footnotes………………………….63
5.1.2 Preserving Idioms and Adding Footnotes…………………………...68
5.2 The Simplification Strategy…………………………………………………74
5.3 Implications of Applying Pre-MT Editing CSIs at the Pragmatic Level.
………………………………………………………………………………….77
Chapter 6 Conclusion………………………………………………….79
6.1 Summary of the Findings……………….…………………………………...79
6.2 Significances of This Research….………….………………………………..82
6.3 Limitations and Suggestions for Future Research….………………………..83
References………………………………………………………………84

I. English
Arnold, D., Balkan, L., Humphreys, R.L., and Sadler, L. (1994). Machine translation.
An introductory guide (pp.210-220). Manchester-Oxford: NCC Blackwell.
Austermuhl, F. (2001). Electronic tools for translators. UK: St. Jerome Publishing.
Baker, M. (1993). Corpus linguistics and translation studies: Implications and
applications. In M. Baker, G. Francis & E. Tognini-Bonelli (Eds.), Text and
technology: In honor of John Sinclair (pp.223-250). Philadelphia & Amsterdam:
John Benjamins.
Baker, M. (1996). Corpus-based translation studies: The challenges that lie ahead. In
H. Somers (Ed.), Terminology, LSP and translation: Studies in language
engineering in honour of Juan C. Sager (pp.175-186). Philadelphia & Amsterdam: John Benjamins.
Brown, P.F., Della Pietra, S.A., Ella Pietra, V.J. & Mercer, R.L. (1990). The mathematics of statistical machine translation: Parameter estimation. Retrieved May, 10th, 2012, from http://acl.ldc.upenn.edu/J/J93-2003.pdf
Burnette, E. ( 2011). Google pulls the rug out from under web service API developers, nixes Google Translate and 17 others. Retrieved April, 8th, 2012, from http://www.zdnet.com/
Carbonell, J.G. & Tomita, M. (1987). Knowledge-based machine translation, the
CMU approaches. In S. Nirenburg (Ed.), Machine translation: Theoretical and
methodological issues (pp.68-89). New York: Cambridge University Press.
Carl, M. (2000). A model of competence for corpus-based Machine Translation. Proceedings of COLING 2000, 2, Germany. Retrieved June, 11th, 2011 from http://aclweb.org/anthology-new/C/C00/C00-2145.pdf
Carl, M. & A. Way. (2003). Introduction. In M. Carl & A. Way (Eds.), Recent advances in example-based machine translation: Text, speech, and language technology (pp. xvii-xxxi). Netherlands: Kluwer Academic Publishers.
EAMT, European Association for Machine Translation. (2012, February 11th). What is
machine translation? Retrieved from http://www.eamt.org/mt.php
Gutt. E.A. (1991). Translation and relevance: Cognition and context. Oxford: Basil
Blackwell.
Google Translate (n.d.). To discover Google Translate. Retrieved May, 16th ,2011, from http://translate.google.com/about/intl/zh-TW_ALL/
Hutchins, J. & Somers, H. (1992). An introduction to machine translation. London: Academic Press.
Hutchins, J. (1986). Machine translation: Past, present, future. Chischester (UK): Ellis Horwood.
Hutchins, J. (1994). Research methods and system designs in machine translation: A ten-year review, 1984-1994. Retrieved October, 1st, 2012 from http://www.hutchinsweb.me.uk/Cranfiled-1994.pdf
Hutchins, J. (1997). Translation technology and the translators. Machine Translation Review, 7, 7-14.
Hutchins, J. (2003). Machine translation: General overview. In R. Mitkov (Ed.), The Oxford handbook of computational linguistics (pp.501-510). New York: Oxford University Press.
Hutchins, J. (2005). The history of machine translation in a nutshell. Retrieved July, 8th, 2012, from http://www.hutchinsweb.me.uk/Nutshell-2005.pdf
Hutchins, J. (2006). Machine translation: History. In K. Brown (Ed.), Encyclopedia of language & linguistics (pp. 370-390). Oxford: Elsevier.
Kaji, H. (1999). Controlled languages for machine translation: State of the art. Retrieved from http://www.mt-archive.info/MTS-1999-Kaji.pdf
Ke, Ping (2009). Machine translation. In M. Baker & G. Saldanha (Eds.), Routledge encyclopedia of translation studies (pp. 160-170). Milton Park, Abidgon: N.Y., NY: Routledge.
Kittredge, R. (2003). Sublanguages and controlled languages. In R. Mitkov (Ed.), The Oxford handbook of computational linguistics. USA: Oxford University.
Lopez, A. (2008). Statistical machine translation. Retrieved March, 13th, 2012, from http://www.erudit.org/~alopez/papers/survey.pdf
Namahn (n.d.). Controlled languages. Retrieved from April 17th, 2012, from http://www.namahn.com/resources/documents/note-CL.pdf
Natural Language Processing. (2011). What is a rule based machine translation system? Retrieved January, 2, 2012, from http://language.worldfcomputing.net/machine-translation/rule-based-machine-translation.html
Och, F. J. (2005). Statistical machine translation: Foundations and recent Advances. In The proceedings of the tenth machine translation summit, Phuket, Thailand, Retrieved December, 26th, 2012, from http://www.mt-archive.info/MTS-2005-Och.pdf
Pym, P.J. (1990). Pre-editing and the use of simplified writing for MT: an engineer’s experience of operating an MT system. In P. Mayocras (Ed.), Translating and the computer 10 (pp. 80-96). London: Aslib.
Quah, C.L. (2006). Translation and technology. New York: Palgrave Macmillan.
Salder, L. (1992). Rule-based translation as constraint resolution. Retrieved May, 9th, 2012, from www.mt-archive.info/FGNLP-1992-Sadler.pdf
Shen, E. (2010). Comparison of online machine translation tools. Retrieved June, 15th, 2012, from www.tcworld.info
Shih, Chung-ling (2006). Helpful assistance to translators: MT & TM. Taipei: Bookman Books.
Shuttleworth, M. and Cowie, M. (1997). Dictionary of translation studies. Manchester, UL.: St. Jerome Publishing.
Smith, K.G. (2000), Bible translation and relevance theory: the translation of Titus. Unpublished doctoral dissertation, University of Stellenbosch.
Sperber, D. and Wilson D. (1995). Relevance: Communication and cognition, (2nd. Ed.) Oxford: Basil Blackwell.
Somers, H. (1998). Machine translation: History. In M. Baker (Ed.), Routledge encyclopedia of translation studies (pp.140-143). London: Routledge.
Somers, H. (2003). Computers and translation, A translator’s guide. Amsterdam & Philadelphia: John Benjamins Publishing Company.
Torrejon, E. and Rico, C. (2002). Controlled translation: a new teaching scenario tailor-made for the translation industry. The proceedings of the 6th EAMT workshop – Teaching machine translation (pp.107-116). Manchester, England: Centre for Computational Linguistics, UMIST.
Melby, Alan K. (1987). On human-machine interaction in translation. In S. Nirenburg (Ed.) Machine Translation (pp. 46-154). Cambridge (UK): Cambridge University Press.
Vanderauwera, R. (1985). Dutch novels translated into English: The translation of a minority literature. Radopi: Amesterdam

II. Chinese
大寶 (1994-2010) ＜白話文、簡化字和現代漢語＞。2012年1月18日取自http://www.edu.cn/20011113/300966.shtml
文學視界 (無日期) ＜語言的源起與種類＞。2012年3月5日取自http://www.white-collar.net
可譯中學中文科 (無日期) ＜語體文（白話文）的源流＞。2012年2月16日取自http://www2.hongai.edu.hk/~ycs/knowledge.doc
史宗玲 (2008)。＜機器翻譯之生產及消費—由解構主義觀點論析＞。《編譯論叢》，第一期，頁113-140。
史宗玲 (2011)。《機器翻譯即時通：台灣籤詩麻ㄟ通》。台北：書林。
百度[Baidu] (2010)。＜民間俗語和成語一樣嗎＞。2012年8月22日取自http://zhidao.baidu.com/question/170112385.html?push=being
余鍵亨 (2005)。《機器翻譯系統為本之雙語網頁對應》國立清華大學資訊工程學
系碩士論文，未出版。
林欣怡 (2011)。《臺灣寺廟觀音靈籤籤詩與解籤文化──以桃園景福宮為例》，
國立臺北教育大學臺灣文化研究所碩士論文，未出版。
林美容 (1998)。＜由地理與年籤來看台灣漢人村莊的命運共同體＞。《台灣風
物》。第38卷4期，頁123-141。
林國平 (2006)。＜籤詩的產生的演變＞。《世界宗教研究》。第四期，頁81-91。
邱崇丙 (2005)。＜俗語界說＞。2012年6月29日取自http://www.yywzw.com/(語
言文字網)
陳鈺淑 (2011)。《屏東縣琉球鄉碧雲寺的籤詩信仰文化研究》，國立屏東教育大
學中國語文學系研究所碩士論文，未出版。
劉玉龍 (2006)。《寺廟籤詩研究-以台灣寺廟運籤為主》，彰化師範大學國文學系
碩士論文，未出版。
劉盛良 (1996)。《機器翻譯中多字動詞問題之研究》，國立台灣大學資訊工程研
究所碩士論文，未出版。
雷赫 (2008)。＜翻譯的“經驗”與“理性”＞，《中國計算機用戶791》。2012年8月
3日取自http://media.ccidnet.com/art/2653/20080105/1343015_1.html
籤詩網（無日期）。＜澎湖天后宮一百籤＞。2011年4月14日取自
http://www.chance.org.tw/