관광 텍스트의 기계번역 포스트에디팅 가이드라인 구축 : 네이버 웨일을 통한 대만 관광 온라인 정보 텍스트 번역 중심으로

다국어 초록 (Multilingual Abstract)

This study proposes a tourism-specific post-editing (PE) guideline for Chinese-Korean machine translation (MT) that aims to improve the readability, accuracy, and cultural appropriateness of automatically translated tourism information. The primary data consist of 300 Chinese sentences drawn from the official tourism websites operated by the Taiwanese local governments of Hsinchu City, Miaoli County, and Changhua County, all of which currently lack Korean-language pages. In this situation, Korean tourists often resort to the browser’s built-in MT function using the Papago engine in the Naver Whale browser for ease of understanding, which means that the quality of MT output directly affects access to local tourism information.
The analysis adopts Kim Hye-rim’s (2022) nine-item Chinese-Korean PE guideline as its basic framework and categorizes errors in terms of accuracy, completeness, consistency, vocabulary, syntactic structure, orthography, punctuation, style, and formatting. On this basis, the study identifies recurring problems such as mistranslated lexical items, unnatural Korean influenced by Chinese comma-based syntax, incomplete transfer of information, and difficulty in rendering cultural references and proper nouns. Drawing on tourism translation theory, the analysis further emphasizes that tourism texts simultaneously perform informational, expressive, and persuasive functions and must therefore introduce local identity and cultural distinctiveness in a way that is acceptable and engaging for foreign visitors. Existing PE guidelines, which were largely developed for general informational or administrative texts, do not fully reflect these functional requirements.
In response, this study refines Kim’s framework by developing PE criteria tailored to tourism discourse, including cultural acceptability, principles for sentence restructuring, and clearer guidance on the treatment of proper nouns, culture-specific items, and reader-oriented expressions. Applying the revised guideline to representative sets of source texts (ST), machine translation (MT), and post-edited output (MTPE) shows that it reduces ambiguity, improves idiomaticity and structural stability, and enhances overall readability. The study thus offers a practical, domain-specific MTPE guideline grounded in authentic Taiwanese tourism data and highlights the need to adapt PE criteria when target texts carry multiple communicative functions and cultural meanings. Limitations include the use of a single post-editor, a relatively small dataset, and reliance on one MT engine; future research should involve multiple post-editors, additional MT systems, and a broader range of tourism and public information texts to test and extend the guideline’s applicability.

목차 (Table of Contents)

• 1. 서론 1
• 1.1. 연구 목적 및 필요성 1
• 1.2. 선행연구 3
• 1.3. 연구대상 및 연구방법 10
• 2. 이론적 배경 12
• 2.1. 기능주의 이론 및 관광 번역 12
• 2.2. 기계번역(MT) 및 포스트에디팅(PE) 16
• 2.3. 관광 텍스트 적용을 위한 오류 양상 검토 24
• 3. 대만 지방 관광 온라인 정보 번역 문제 유형 및 논의· 28
• 3.1. 기능주의 이론 및 관광 번역 28
• 3.2. 주요 수정 양상 및 논의 31
• 4. 관광 텍스트에 대한 수정된 포스트에디팅 가이드라인의 적용 분석· 41
• 4.1. 정확성 42
• 4.2. 완결성 50
• 4.3. 문화 수용성 54
• 4.4. 어휘 62
• 4.5. 통사구조 71
• 4.6. 맞춤법 78
• 4.7. 문장부호 82
• 4.8. 문체 84
• 5. 결론 90
• 참고문헌 92
• 부록 96
• ABSTRACT 180