ChatGPT与常用翻译软件的对比研究——以医学文本为例  

A Comparative Study of ChatGPT and Commonly Used Translation Software——A Case Study with Medical Texts

作　　者：张晓[1] 杨占[1] Zhang Xiao;Yang Zhan(Chongqing Medical University,Chongqing,400016,China)

机构地区：[1]重庆医科大学,重庆400016

出　　处：《语言与文化研究》2024年第5期195-198,共4页Language and Culture Research

摘　　要：ChatGPT是由OpenAI开发的一种自然语言处理模型,能够理解和生成自然语言文本,可用于多种自然语言处理任务。本文对当前热门的ChatGPT与常用的国内外翻译软件进行比较,并从词汇翻译和表达规范的角度进行分析。结果表明,无论是常用机器翻译软件还是ChatGPT,在医学专业术语的翻译方面表现出较高的准确性,能够相对准确地传达医学术语的含义。然而,但在医学两栖词汇的翻译上仍然存在一定的挑战,难以完全实现医学语境下的准确翻译。此外,这四款软件在医学文本的规范表达方面仍有改进的空间。但与常用的翻译软件相比,ChatGPT具备优化译文的能力,可以更好地调整翻译结果,使之更贴近医学文本的语言风格。ChatGPT is a natural language processing model developed by OpenAI,capable of comprehending and generating natural language text,serving various natural language processing tasks.This article conducts a comparative study between ChatGPT and widely employed domestic and foreign translation software,utilizing medical texts as a case study.The comparison is centered around vocabulary translation and adherence to expression standards.The findings indicate that both common machine translation software and ChatGPT exhibit a high degree of accuracy in translating medical terminology,effectively conveying the intended meaning.Nevertheless,challenges persist in translating ambiguous medical vocabulary,and achieving complete precision within a medical context remains elusive.Furthermore,there is room for enhancement in the standardized expression of medical texts for all five software applications.In contrast,ChatGPT demonstrates the capability to optimize translations and adjust results more effectively to align with the linguistic style of medical texts,setting it apart from conventional translation software.

关 键 词：人工智能 ChatGPT 医学文本 机器翻译 

分 类 号：H31[语言文字—英语]