摘要
神经网络机器翻译技术的快速发展极大地提升了翻译的准确性和效率，机器翻译在各专业领域得到广泛应用。然而，受限于语料、专业知识与语境理解能力，机器翻译在处理航空文本时仍存在诸多问题，尤以词汇层面和句法层面最为突出。本研究以航空文本为研究对象，旨在探讨机器翻译中常见的错误类型，并借助译后编辑有效提升译文的专业性、可读性和整体翻译质量。研究发现，词汇层面的术语误译与语境不适配容易导致概念混淆甚至信息错误，句法层面的照搬原文结构和语义模糊问题，使译文逻辑混乱、表达生硬，影响译文的可读性与信息传达的准确性。译后编辑能够准确识别并改正词汇层面和句法层面错误，提升译文准确性与可读性。希望本文为同类型文本译后编辑提供有益的参考和借鉴，推动航空领域机器翻译译后编辑的进一步探索与实践。
关键词： 航空文本；机器翻译；问题类型；译后编辑
Abstract
The rapid development of neural machine translation (NMT) technology has significantly improved translation accuracy and efficiency, leading to its widespread application across various specialized fields. However, due to limitations in corpus data, specialized knowledge, and contextual understanding, machine translation still encounters numerous challenges in translating aviation texts, with lexical and syntactic issues being particularly prominent. This study focuses on aviation texts and aims to identify common error types in machine translation. By applying pose-editing, it seeks to effectively enhance the professionalism, readability, and overall quality of the output. It is found that terminological mistranslations and contextual mismatches at the lexical level are likely to cause conceptual confusion and even factual errors. At the syntactic level, rigid adherence to the source structure and semantic ambiguity result in disorganized logic and awkward expression, undermining the readability of the translation and the accuracy of information delivery. Post-editing can effectively identify and correct errors at both the lexical and syntactic levels, thereby improving the accuracy and readability of the translation. It is hoped that this study will provide valuable reference for post-editing similar texts and promote the further exploration and practice of machine translation post-editing in the aviation field.
Key words： Aviation texts; Machine translation; Types of errors; Post-Editing
参考文献 References
[1] Allen, J. Posting-editing [A]. In Somers, H. (ed). Computers and Translation: A Translator’s Guide [M]. John Benjamins Publishing, Amsterdam, 2003.
[2] Daems, Macken L, Vandepitte S. Quality as the sum of its parts: a two-step approach for the identification of translation problems and translation quality assessment for HT and MT+PE [C]//Proceedings of MT Summit XIV Workshop on Post-Editing Technology and Practice, 2013, 2:63-71.
[3] DePalma, D. A. Post-edited machine translation define [EB/OL]. (2013-04-30)[2014-02-21].
[4] ISO 18587, Translation services — Post-editing of machine translation output — Requirements [S], 2014.
[5] Krings, H. P. Repairing Texts: Empirical Investigations of Machine Translation Post-editing Processes [M]. Kent: Kent State University Press, 2001.
[6] Tezan A. Informative quality estimation of machine translation output [D]. Ghent University, 2018.
[7] Vilar D, Xu J, Luis Fernando D H, et al. Error analysis of statistical machine translation output [C]// 5th International Conference on Language Resources and Evaluation, 2006: 697-702.
[8] 崔启亮.论机器翻译的译后编辑[J].中国翻译, 2014, (6):68-73.
[9] 胡家敏,冷冰冰.MTPE模式下科技文本典型翻译错误的归类与分析[J].中国科技翻译, 2023, 36(01): 23-26.
[10] 罗季美,李梅.机器翻译译文错误分析[J].中国翻译, 2012, 33(05): 84-89.
[11] 孟福永,唐旭日.效率为先：机器翻译译后编辑技术综述[J].计算机工程与应用, 2020, 56(22): 25-32.
[12] 王华树、刘世界. 人工智能时代机器翻译译后编辑能力探究[J].中国科技翻译, 2022, (4): 21-24.
[13] 杨文地,范梓锐.科技语篇机器翻译的译后编辑例析[J].上海翻译, 2021, (06): 54-59.
[14] 姚斌.联合国文件机器翻译及译后编辑问题探析[J].中国翻译, 2024, (4): 144-151.
[15] 于璇,冷冰冰.医疗器械文本译后编辑研究——以呼吸机用户手册汉译为例[J].中国科技翻译, 2025, 38(02): 22-25.