번역사의 번역교육 경험이 포스트에디팅 결과물의 품질에 미치는 영향에 관한 사례연구
A Case Study on the Influence of Translator’s Experience in Translation Education on the Quality of Post-Editing Results
1
조회
0
다운로드
20
인용
0
참고문헌
키워드
초록
The purpose of this study is to identify the impact of traditional translation education on the quality of MTPE results. To this end, the study analyzed the MTPE results of three translators with different levels of translation education. The texts for analysis were selected from the National Information Society Agency's AI data creation project, which built a parallel corpus for Korean-English machine translation. Post-editing guidelines used in the project were used as basis for analysis on which the number and severity of errors generated were analyzed. According to the analysis, translator A, who graduated from a graduate program in interpretation and translation, showed the most errors. Translator C, who does not have any translation background, and translator B, who finished a master’s program in translation, followed in order. Translator A's style of work seems to have affected this result, which showed a rapid pace of work but missed many errors in machine translation. Serious errors appeared in the order of translator C, translator A, and translator B. In the meantime, by subject, there was a difference in the quality of the results of all post-editors. These results mean that translation education can have some effect on MTPE quality. Based on the analysis results, the subject-specific translation practice highlighted in the existing translation training can help to improve the quality differences that arise according to topic. In addition, the addition of editing competencies and CAT tool exercises to MTPE training will contribute to improving MTPE quality and improving the expertise of translators. The study has a limit to the fact that the quality difference was not clearly identified because individual factors affecting post-editing quality, such as the working style of translators and subject areas familiar to each translator, were not sufficiently excluded. However, it is urgent to establish the direction of translation education in order to respond to the changing situation of the ‘coming future’, and the research is meaningful, especially in that the study aimed to find out the direction of traditional translation education and post-editing curriculum by utilizing the results of post-editing at the actual industrial site.
Abstract
The purpose of this study is to identify the impact of traditional translation education on the quality of MTPE results. To this end, the study analyzed the MTPE results of three translators with different levels of translation education. The texts for analysis were selected from the National Information Society Agency's AI data creation project, which built a parallel corpus for Korean-English machine translation. Post-editing guidelines used in the project were used as basis for analysis on which the number and severity of errors generated were analyzed. According to the analysis, translator A, who graduated from a graduate program in interpretation and translation, showed the most errors. Translator C, who does not have any translation background, and translator B, who finished a master’s program in translation, followed in order. Translator A's style of work seems to have affected this result, which showed a rapid pace of work but missed many errors in machine translation. Serious errors appeared in the order of translator C, translator A, and translator B. In the meantime, by subject, there was a difference in the quality of the results of all post-editors. These results mean that translation education can have some effect on MTPE quality. Based on the analysis results, the subject-specific translation practice highlighted in the existing translation training can help to improve the quality differences that arise according to topic. In addition, the addition of editing competencies and CAT tool exercises to MTPE training will contribute to improving MTPE quality and improving the expertise of translators. The study has a limit to the fact that the quality difference was not clearly identified because individual factors affecting post-editing quality, such as the working style of translators and subject areas familiar to each translator, were not sufficiently excluded. However, it is urgent to establish the direction of translation education in order to respond to the changing situation of the ‘coming future’, and the research is meaningful, especially in that the study aimed to find out the direction of traditional translation education and post-editing curriculum by utilizing the results of post-editing at the actual industrial site.
참고문헌 (27)
- 김순미, "번역학계와 언어서비스업체(LSP)간 산학협력연구: ‘포스트에디팅 생산성’과 ‘기계번역 엔진 성능 비교’", 번역학연구, 20(1), pp.41~76 (2019), 한국번역학회[DOI]
- 김순영, "학부 번역교육의 효율성 제고방안에 관한 소고: 이론 기반 텍스트 분석 교과의 도입", 통역과 번역, 8(1), pp.27~43 (2006), 한국통역번역학회
- 김순영, "국내 학부 번역교육과정에 대한 고찰 —번역능력(translation competence)의 관점에서", 통번역교육연구, 6(2), pp.5~22 (2008), 한국통번역교육학회
- 남보라, "2019 NIA 자동번역후편집(MTPE) 가이드라인" (2019), ㈜솔트룩스파 트너스
- 박혜경, "석사 과정의 기계번역 수업에 대한 소고: 한일번역 전공생의 포스트에디팅 사례를 통하여", 번역학연구, 19(3), pp.163~193 (2018), 한국번역학회[DOI]
- 서보현, "기계번역 결과물의 오류유형 고찰", 번역학연구, 19(1), pp.99~117 (2018), 한국번역학회[DOI]
- 신석환, "한국어·영어 번역 말뭉치 AI데이터 구축 가이드라인" (2019), 한국정보화진흥원
- 신지선, "번역교육 관련 연구동향 및 연구과제 고찰-교육내용, 수업방법, 평가방식을 중심으로", 통역과 번역, 15(2), pp.131~150 (2013), 한국통역번역학회
- 신지선, "포스트에디팅: 최적의 협업방식을 찾아서", KIGO 소식지, 7, pp.38~53 (2017)
- 신지선, "번역 교육의 새로운 ‘통합’ 패러다임 제안", 통번역교육연구, 17(2), pp.43~62 (2019), 한국통번역교육학회[DOI]
- 이상빈, "학부번역전공자의 기계번역 포스트에디팅, 무엇이 문제이고, 무엇을 가르쳐야 하는가?", 통역과 번역, 19(3), pp.37~64 (2017), 한국통역번역학회[DOI]
- Forcada, Mikel, "Handbook of translation studies 1", pp.215~223 (2010), John Benjamins
- Amparo, Hurtado Albir, "Competence-Based Curriculum Design for Training Translators", The Interpreter and Translator Trainer, 1(2), pp.163~195 (2007)
- Jia, Yanfang, "How Does the Post-Editing of Neural Machine Translation Compare with From-Scratch Translation? A Product and Process Study", The Journal of Speciallised Translation, 31, pp.60~86 (2019)
- Krings, Hans Peter, "Reparing Texts: Empirical Investigations of Machine Translation Post-Editing Processes" (2001), The Kent State UP
- Läubli, Samuel, "Post-editing Productivity with Neural Machine Translation: An Empirical Assessment of Speed and Quality in the Banking and Finance Domain", Proceedings of Machine Translation Summit XVII, 1, pp.267~272 (2019)
- O’Brien, Sharon, "Teaching Post-Editing: a Proposal for Course Content", Proceedings of the Sixth EAMT Workshop, pp.99~106 (2002)
- Robert, Isabelle, "Conceptualizing Translation Revision Competence: A Pilot Study on the ‘Fairness and Tolerance’ Attitudinal Component", Perspectives, 26(1), pp.2~23 (2017)
- Sager, Juan C, "Language Engineering and Translation: Consequences of Automation" (1994), John Benjamins
- Yamada, Masaru, "The Impact of Google Neural Machine Translation on Post-Editing by Student Translators", The Journal of Speciallised Translation, 31, pp.87~106 (2019)
- "인간 vs AI 번역대결서 인간 압승… AI, 문학 번역 취약" (2017), 윤동지[URL]
- "한국어-영어 번역(병렬) 말뭉치 AI데이터" (2019), 한국정보화진흥원[URL]
- "Evaluating models" (2020)[URL]
- "ISO 17100:2015 Translation services - requirement for translation services" (2015)[URL]
- "ISO 18587:2017 Translation services - post-editing of machine translation output – requirements" (2017)[URL]
- "The Fourth Industrial Revolution: What it means and how to respond" (2015)[URL]
- "ound in translation: more accurate, fluent sentences in Google Translate" (2016)[URL]