‘ 구글 번역 ’ 기반 텍스트 유형별 중한 기계번역오류 현황 분석
Analysis of Errors in Google Chinese-Korean Machine Translation System
0
조회
0
다운로드
14
인용
0
참고문헌
키워드
초록
The purpose of this study is to induce the types of error shown in the performances of Google machine translation system with the three types of text such as idioms, conversations, and articles. 85 idioms out of 100 idioms were not correctly translated. As a result of analyzing the error patterns concerning the 85 idioms, six types of errors such as omission of essential components, excessive analogy, misused vocabularies, marking of pronunciations, fail to embody event structures, and misused proper nouns were deduced. The dialogue had errors in 30 sentences out of 127 sentences. The causes of these errors such as omission of essential components, misused proper nouns, misused vocabularies, and unrecognized idiomatic expressions can be deduced. 59 errors were extracted from the articles. These errors can be categorized into seven types: misused proper nouns, omission of essential components, misused vocabularies, abuse of cultural elements, fail to embody syntactic meanings, fail to embody logical structures, and failed to apply article types.
Abstract
The purpose of this study is to induce the types of error shown in the performances of Google machine translation system with the three types of text such as idioms, conversations, and articles. 85 idioms out of 100 idioms were not correctly translated. As a result of analyzing the error patterns concerning the 85 idioms, six types of errors such as omission of essential components, excessive analogy, misused vocabularies, marking of pronunciations, fail to embody event structures, and misused proper nouns were deduced. The dialogue had errors in 30 sentences out of 127 sentences. The causes of these errors such as omission of essential components, misused proper nouns, misused vocabularies, and unrecognized idiomatic expressions can be deduced. 59 errors were extracted from the articles. These errors can be categorized into seven types: misused proper nouns, omission of essential components, misused vocabularies, abuse of cultural elements, fail to embody syntactic meanings, fail to embody logical structures, and failed to apply article types.
참고문헌 (15)
- 강병규, "신경망 기계번역의 작동 원리와 번역의 정확률 - 중한 번역을 실례로", 중어중문학, (73), pp.253~295 (2018), 한국중어중문학회[DOI]
- 기유미, "한중 기계번역 오류의 문형별 비교분석-네이버 파파고 번역기와 구글 번역기를 중심으로-", 중국연구, 74, pp.3~32 (2018), 중국연구소[DOI]
- 김보름, "중국어뱅크 베이시스국어(STEP 1)" (2017), 동양 북스
- 김운, "중한 대화체 자동번역을 위한 중국어 긴축문 처리", 한국 정보과학회 학술발표논문집, 39, pp.300~302 (2012)
- 박재승, "중한 기계번역에서의 给자문 처리기제 연구", 중국어교육과 연구, (9), pp.101~113 (2009), 한국중국어교육학회
- 배재석, "중한 기계 번역 시스템에서의 오류 분석", 중국언어연구, (18), pp.341~369 (2004), 한국중국언어학회
- 송인성, "中韓 기계 번역을 위한 中國語 單語 自動 분리", 중국언어연구, 14, pp.69~92 (2002), 한국중국언어학회
- 신경미, "중국어 자동 번역 시스템 현황과 시스템 구축을 위한 제안", 한중인문학연구, (56), pp.199~219 (2017), 한중인문학회[DOI]
- 양재형, "중한번역에서 구묶음을 이용한 파싱 효율 개선", 정보과학회논문지 : 소프트웨어 및 응용, 31(8), pp.1083~1091 (2004), 한국정보과학회
- 윤성희, "다중 엔진을 이용한 중-한 기계번역 시스템의 설계", 공학기술연구, 2004(3), pp.1~6 (2004)
- 尹斌庸, "博古通今学汉语丛书-成语100" (1999), 华语教学出版社
- "韩国央行决定维持基准利率不变"[URL]
- "韩国总统文在寅改组内阁 拟更换国防部长"[URL]
- "구글 번역"[URL]