翻译质量评估的现状与对策:基于人文社科与自然科学文献的计量研究(1981—2021)

Problems and Prospects of Translation Quality Assessment:A Bibliometric Review Based on Research Articles from the Humanity&Social Science and Natural Science Disciplines(1981—2021)

随着机器翻译技术的不断进步,翻译质量评估模式发生了深刻的变化,评估理论与评测技术有了新的进展。以全新的视角回顾人工评估与机器翻译自动评测所取得的成果对ESP翻译研究的发展具有重要的理论和实际意义。在综述了基于N-gram相似度、翻译编辑率、术语质量和向量相似度等评估指标的基础上,本文创新性地引入基于Python统计与词向量聚类的量化实验。结果显示,由于翻译质量评估涉及不同学科,其研究往往呈现零散、孤立等特点。通过对比研究可以发现,该领域学者分别来自人文社科与自然科学,研究主题与范式大相径庭,呈现出各自为政的研究态势。针对当前研究存在的问题,本文提出相关建议,以期为ESP翻译教育与评价的跨学科研究提供参考。

With the advancement of machine translation technology,translation quality assessment has undergone profound changes.Examining human and automatic assessment of translation from a new perspective is important for the research and practice of Translation of English for Specific Purposes.Bringing together history and recent developments in human/machine translation quality assessment(TQA),the study aims to present a comprehensive overview of the field.Drawing upon 1575 research articles,this study begins by reviewing previous research(including N-gram metrics,Translation Edit Rate,Terminology-based evaluation metrics,and vector-based metrics),proceeds to introduce the state-of-the-art in the field of TQA,and implements word vectors and k-means clustering through Python for data analysis.Major findings include:1)due to the interdisciplinary nature of the field,the studies of TQA tend to be disconnected and fragmented;2)a comparative study reveals that scholars from different disciplines tend to focus on different research topics and adopt varied research paradigms.These findings open up new directions for future research and provide implications for the practice of translation quality assessment in the field of ESP.