기계번역과 인간번역, 기계번역 사후편집본과 인간번역 감수본의 품질평가 비교 연구 - 한-노 번역 사례를 중심으로
A Comparative Evaluation of Machine Translation and Human Translation, and Post-edited Machine Translation and Revised Human Translation: Focusing on Korean-Russian Translation
0
조회
0
다운로드
9
인용
0
참고문헌
키워드
초록
This study compares and analyzes machine and human translation, then post-edited machine translation and revised human translation. Source texts were composed of colloquial and written styles, respectively, since the results of MT and MTPE may vary. Evaluation and qualitative assessment were manually undertaken based on the TQA criteria proposed by Han (2021). Professional translators codded and scored errors, and Korean-Russsian machine translation and human translation differ in frequency and distribution of errors. The form-related 'accuracy' errors were distinctive in machine translation. In contrast, errors in meaning-related 'clarity/comprehensibility' occurred more with human translation. After machine translation is post-edited and human translation is revised, the quality of each output was significantly improved; however, some error types somewhat increased despite post-editing and revision efforts. Reviewers tended to assess human translators' clarification and explication attempts as excessive intervention. Moreover, post-edited machine translation and revised human translation showed little difference in evaluation scores, indicating that specialized training for post-editors (or reviewers) will inevitably enhance machine translation quality.
Abstract
This study compares and analyzes machine and human translation, then post-edited machine translation and revised human translation. Source texts were composed of colloquial and written styles, respectively, since the results of MT and MTPE may vary. Evaluation and qualitative assessment were manually undertaken based on the TQA criteria proposed by Han (2021). Professional translators codded and scored errors, and Korean-Russsian machine translation and human translation differ in frequency and distribution of errors. The form-related 'accuracy' errors were distinctive in machine translation. In contrast, errors in meaning-related 'clarity/comprehensibility' occurred more with human translation. After machine translation is post-edited and human translation is revised, the quality of each output was significantly improved; however, some error types somewhat increased despite post-editing and revision efforts. Reviewers tended to assess human translators' clarification and explication attempts as excessive intervention. Moreover, post-edited machine translation and revised human translation showed little difference in evaluation scores, indicating that specialized training for post-editors (or reviewers) will inevitably enhance machine translation quality.
참고문헌 (31)
- 곽중철, "포스트에디팅 측정지표를 통한 기계번역 오류 유형화 연구", 통번역학연구, 22(1), pp.1~25 (2018), 통번역연구소[DOI]
- 정혜연, "번역자동평가에서 풀리지 않은 과제", 번역학연구, 21(1), pp.9~29 (2020), 한국번역학회[DOI]
- 김순미, "AI시대 인간번역과 기계(NMT)번역의 공존-경영학 ‘확장(Augmentation)전략’ 중심", 통역과 번역, 20(2), pp.1~32 (2018), 한국통역번역학회[DOI]
- 김순미, "번역학계와 언어서비스업체(LSP)간 산학협력연구: ‘포스트에디팅 생산성’과 ‘기계번역 엔진 성능 비교’", 번역학연구, 20(1), pp.41~76 (2019), 한국번역학회[DOI]
- 김진아, "번역 품질평가에 관한 小考", 외국문학연구, (11), pp.4~124 (2002), 외국문학연구소
- 노경섭, "제대로 알고 쓰는 논문 통계분석: SPSS & AMOS 21" (2014), 한빛아카데미
- 박지영, "감수자 자격과 감수품질에 관한 제언: 한영번역에 대한 원어민 감수를 중심으로", 통역과 번역, 22(3), pp.1~27 (2020), 한국통역번역학회[DOI]
- 박지영, "원어민 감수자의 자격, 능력, 직무에 관한 고찰", 번역학연구, 22(3), pp.137~164 (2021), 한국번역학회[DOI]
- 서보현, "기계번역 결과물의 오류유형 고찰", 번역학연구, 19(1), pp.99~117 (2018), 한국번역학회[DOI]
- 신지선, "기계번역 포스트에디팅에 관한 해외 연구 동향", 번역학연구, 21(4), pp.87~114 (2020), 한국번역학회[DOI]
- 이상모, "기계번역을 활용한 법령번역의 실제와 과제", T&I REVIEW, 11(1), pp.35~56 (2021), 통역번역연구소[DOI]
- 이주리애, "인공신경망 기계 번역의 한일", 일한 번역 품질에 대한 예비연구-품질향상 검토와 교열코드 적용-, 2018, pp.20(1), 43~71[DOI]
- 이향, "번역품질평가의 현황과 과제", 번역학연구, 11(2), pp.107~125 (2010), 한국번역학회[DOI]
- 전혜진, "텍스트 유형별 러한 기계번역 포스트에디팅 연구", 슬라브어 연구, 24(2), pp.247~274 (2019), 한국슬라브어학회[DOI]
- 정혜연, "번역의 자동평가: 기계번역 평가를 인간번역 평가에 적용해보기", 통번역학연구, 22(4), pp.265~287 (2018), 통번역연구소[DOI]
- 한승희, "기계번역·인간번역·트랜스크리에이션의 문체 비교: 광고 번역을 중심으로", 통번역학연구, 21(2), pp.163~188 (2017), 통번역연구소[DOI]
- 한승희, "기계번역 품질에 대한 요인분석", 통번역학연구, 25(2), pp.147~170 (2021), 통번역연구소[DOI]
- 황지연, "한국어-중국어 신경망 기계 번역에 대한 수동평가: 사회과학과 기술과학 카테고리에서 나타나는 구글 번역(GNMT)과 파파고(N2MT)의 정확도", 통번역학연구, 24(4), pp.175~206 (2020), 통번역연구소[DOI]
- 한현희, "한-노 기계 번역, 어디까지 왔나?:Google과 Papago 번역 성능 비교를 기반으로", 노어노문학, 32(3), pp.63~94 (2020), 한국노어노문학회
- Arnold, D., "Machine translation: An introductory guide" (1994), Blackwell Pub
- Campbell, S., "Translation into the second language" (2018), Routledge
- Doherty, S., "The Impact of Translation Technologies on the Process and Product of Translation", International Journal of Communication, 10, pp.947~969 (2016)
- Dyson, M. C., "Toward a methodology for the evaluation of machine-assisted translation systems", Computers and translation, 2(2-3), pp.163~176 (1987)
- Floyd, J. F., "Factor analysis in the development and refinement of clinical assessment instruments", Psychological Assessment, 7(3), pp.286~299 (1995)
- Gorsuch, R. L., "Exploratory factor analysis : Its role in item Analysis", Journal of Personality Assessment, 68(3), pp.532~560 (1997)
- Larose, R., "Méthodologie de l'évaluation des traductions", Meta : journal des traducteurs, 1998, pp.43(2), 163~186
- Mossop, B., "번역가를 위한 편집과 교정" (2011), 동인
- Павлова, А. В., "Оценка качества перевода. Вестник Пермскогонационального исследовательского политехническогоуниверситета", Проблемы языкознания и педагогики, 6(45) (2012)
- Переходько, И. В., "Оценка качествакомпьютерного перевода", Вестник Оренбургскогогосударственного университета, 2(202) (2017)
- Сдобников, В. В., "Оценка качества перевода(коммуникативно-функциональный подход)" (2015)
- "“5도 사투리 음성·10개 암 영상…AI 학습용데이터 170종 풀린다”" (2021)[URL]