A Study On Pre-editing Strategies In Chinese To English Machine Translation Of IT And Communications Texts

In recent years,machine translation(MT)has been widely employed in various fields due to the rapid development of MT technology,the continuously improved MT output quality,and the increasing applicability of MT systems to different types of texts.As an important approach to improve translation efficiency,MT plays a key role in meeting the great demand for translation of IT and communications texts in translation markets since the MT quality of IT and communications texts is relatively high.Based on the author's translation practice during an internship in a communications technology company,this thesis analyzes Chinese to English(CE)MT of IT and communications texts.After the quality evaluation and error analysis,this thesis first summarizes nine error types existing in the studied corpus,including mistranslation,omission,redundancy,mistranslation of terminology,typographical errors in terminology,grammatical errors,violation of the manual of style,inappropriate translation,and wrong collocation.Then,to avoid the appearance of the preceding errors,this thesis proposes several corresponding pre-editing strategies,such as preprocessing GUI strings or terminologies,adjusting sentence structure,as well as eliminating ambiguity by supplementing necessary information,clarifying logical relation,deleting redundant information,or utilizing punctuations.In addition,the effectiveness of the proposed strategies is verified during case analysis.This thesis is conducive to assisting translators in optimizing source texts by applying pre-editing strategies to avoid MT output errors,improve MT quality,and reduce the workloads for post-editing,thus improving translation efficiency and meeting the market demand.