Machine Translation Outputs: Korean to English Translation of Embedded Sentences
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
The recent machine translation has been operated by two main models: the Neural Machine Translation model (NMT) and the Statistical Machine Translation model (SMT). The present study is to give a better understanding of the machine translation for the users and to provide suggestions about what further improvement would be required. In this work, the analysis of the Korean-to-English translation errors in two machines is accomplished by using two movie scripts. Most of all, we investigate which type of embedded clauses make machine translation more problematic in reference to a level of embedding. The results indicate that NMT has better outputs than SMT does in terms of its naturalness. However, when the NMT doesn’t have a corresponding sentence in its corpus data, it could make more serious semantic errors than the SMT, making either non-sensical sentences or considerably far-fetched sentences.
Abstract
The recent machine translation has been operated by two main models: the Neural Machine Translation model (NMT) and the Statistical Machine Translation model (SMT). The present study is to give a better understanding of the machine translation for the users and to provide suggestions about what further improvement would be required. In this work, the analysis of the Korean-to-English translation errors in two machines is accomplished by using two movie scripts. Most of all, we investigate which type of embedded clauses make machine translation more problematic in reference to a level of embedding. The results indicate that NMT has better outputs than SMT does in terms of its naturalness. However, when the NMT doesn’t have a corresponding sentence in its corpus data, it could make more serious semantic errors than the SMT, making either non-sensical sentences or considerably far-fetched sentences.
참고문헌 (29)
- 배지연, "한영 기계 번역에 있어서 격 실현 현상 고찰 -격 중출 구문을 중심으로-", 언어과학, 17(4), pp.99~120 (2010), 한국언어과학회
- Callison-Burch, Chris, "Re-evaluation the role of Bleu in machine translation research", EACL, 6, pp.249~256 (2006)
- 최동익, "과거시제의 영어 복문과 중문에 대한 번역가 번역과 기계번역 비교", 언어학 연구, (39), pp.377~398 (2016), 한국중원언어학회[DOI]
- 최효은, "특허 기계번역 결과물의 평가 - KIPRIS의 무료 한영 기계번역을 중심으로", 통역과 번역, 19(1), pp.139~178 (2017), 한국통역번역학회
- Doddington, George, "Automatic evaluation of machine translation quality using n-gram co-occurrence statistics", Proceedings of the Second International Conference on Human Language Technology Research (HLT-02), pp.138~145 (2002)
- 동연, "한국어 내포문의 중국어 대응문장 고찰", 이화어문논집, (38), pp.109~132 (2016), 이화어문학회
- Ghasemi, Hadis, "A comparative study of Google translate translations: An error analysis of English-to-Persian and Persian-to-English translations", English Language Teaching, 9(3), pp.13~17 (2016)
- Guzmán, Francisco, "Machine translation evaluation with neural networks", Computer Speech and Language, 45, pp.180~200 (2017)
- 정경미, "관형사절의 한영 기계번역을 위한 복합문 분절 방법에 대한 연구", 우리말글, 49, pp.37~62 (2010), 우리말글학회
- 김경석, "번역 오류 분석: 어휘를 중심으로", 새한영어영문학, 48(3), pp.113~136 (2006), 새한영어영문학회
- Kwon, Jae-il, "Korean grammar" (2012), Thaehaksa
- Lee, Changki, "English-to-Japanese machine translation using neural networks", Communications of the Korean Institute of Information Scientists and Engineer, 33(10), pp.48~52 (2015)
- 이현영, "내포문의 단문 분할을 이용한 한국어 구문 분석", 정보과학회논문지 : 소프트웨어 및 응용, 35(1), pp.50~58 (2008), 한국정보과학회
- Lee, Yong-hun, "Processing noun-modifying clauses in Korean-English machine translation", Proceedings of Annual Conference on Human and Cognitive Language Technology, 16(1), pp.232~239 (2004)
- Lee, Young-Ok, "Problems of Korean-English translation: With respect to passive constructions", The Journal of Translation Studies, 1(2), pp.47~76 (2000)
- 임순정, "구글 번역의 사례연구: 번역품질과 번역단위와의 관계를 중심으로", 번역학연구, 15(1), pp.177~209 (2014), 한국번역학회
- Nida, Eugene Albert, "The theory and practice of translation" (1982), Brill
- Oh, Mihyung, "A study on the English subtitling strategies in Korean wordplay", Ph. D, pp.2010
- Papineni, Kishore, "BLEU: A method for automatic evaluation of machine translation", Proceedings of the 40th Annual Meeting on Association for Computational Linguistics (ACL-02), pp.311~318 (2002)
- 박현주, "공공번역 결과물 오류 유형 분석- 공공기관 영어 웹사이트를 중심으로", 통역과 번역, 16(1), pp.51~80 (2014), 한국통역번역학회
- 박옥수, "한영문학 번역의 대화체에서 드러난 생략어의 번역", 동화와 번역, (25), pp.171~192 (2013), 동화와번역연구소
- 박옥수, "관형절의 한영 번역 기법", 동아인문학, (33), pp.285~307 (2015), 동아인문학회
- Park, Jaehong, "Comparative analysis of word", subword, Proceedings of Symposium of the Korean Institute of Communication Sciences, pp.2016[DOI]
- 설옥순, "시제와 시상에 대한 영한 번역 전략", 번역학연구, 10(2), pp.23~60 (2009), 한국번역학회
- Suh, Sungki, "Center-embedding and parsing breakdown in English and Korean", Korean Journal of Applied Linguistics, 16(2), pp.127~148 (2000)
- 원은하, "영한번역에서 구조적 중의성의 처리 전략: 내포문 구성을 중심으로", 번역학연구, 13(1), pp.87~114 (2012), 한국번역학회
- Yang, Seong-il, "Design of verb-phrase patterns for Korean to English MT", Proceedings of Korea Information Science Society, 28(2), pp.178~180 (2001)
- YomHaeng-il, "번역의 오류 분석 -한영번역을 중심으로-", 한국언어문화, 24, pp.77~92 (2003), 한국언어문화학회
- "Deep Learning and Translation by NAVER CAST"[URL]