Research On Error Analysis Of English-chinese Machine Translation Of Business English And Countermeasures

At present,few academic papers explored error analysis in machine translation(MT)of Business English into Chinese.Present studies are limited on translation error classifications only.Machine translation(MT)has developed rapidly in recent years.It enjoys unique advantages with its high translating speed and free access compared with manual translation.The constant improvement of MT quality can meet the basic needs of rough reading but when it comes to the needs of intensive reading,the quality of MT output needs to be further improved.Based on analysis of articles from the wall street journal,this paper aims to improve the quality of English-Chinese MT by conducting an error analysis.This paper analyzed the types of errors in MT translation,and discussed possible attributing factors for error distribution.Moreover,this paper compares MT outputs of two-type texts from the perspective of the naming-telling clause(NT clause,hereafter).First,two types of the same text were processed as input into an MT engine,and their output are compared hereafter.One type of input is the original English sentence;the other type of input is the same texts,but all sentences are processed based on NT clause structure.The errors of MT output are categorized into three classes: incorrect lexical choice,structural errors and component omissions.Structural errors are further divided into SV structural errors and non-SV structural errors.The analyzed data shows,firstly,the major errors are structural errors,in which non-SV-structural errors account for a larger proportion;secondly,translation errors decrease significantly after English sentences are parsed into NT clauses.This result reveals that non-SV clauses are the main source of MT errors,and suggests that English sentences,especially for long English sentences,should be parsed into NT clauses before they are translated for better translation quality.