基于GLR算法的英语专用词汇翻译智能检错模型研究  

Intelligent Error Detection Model for English Specialized Vocabulary Translation Based on GLR Algorithm

作　　者：吴娜 Wu Na(School of International Business,Shenyang Normal University,Shenyang 110034,China)

机构地区：[1]沈阳师范大学国际商学院,辽宁沈阳110034

出　　处：《黑河学院学报》2024年第10期112-114,共3页Journal of Heihe University

基　　金：教育部产学研合作协同育人项目“‘互联网’背景下依托企业多维协同培养商科师资队伍建设”(220600228204753)。

摘　　要：英语词汇量庞大,专用词汇更是层出不穷,涵盖专业术语、行业术语、新创词汇等,这些词汇的翻译检错具有较大难度。建立基于广义最大似然比检测(GLR)算法的英语专用词汇翻译智能检错模型。基于GLR算法计算各类英语专用词汇翻译错误的可分离性,选择可分离性高的文本项作为各种翻译检错的特征文本项,并构建英语专用词汇翻译错误检测模型。实验结果表显示:研究所构建模型对于通用领域翻译文本、特定领域翻译文本、口语化翻译文本三类英语专用词汇翻译的检错,均具有高于95%的准确率,这显著优于传统方法。以上测试结果说明该模型的应用可靠性更高。English has a huge vocabulary with constantly emerging specialized vocabulary including professional terminology,industry terminology,and innovative vocabulary.It’s quite diffi cult to translate and detecting the error of these vocabulary items.To this end,it’s necessary to construct an intelligent error detection model for English specialized vocabulary translation based on the generalized maximum likelihood ratio detection(GLR)algorithm.In this study,the GLR algorithm is used to calculate the separability of various errors in English specialized vocabulary translation,and text items with high separability are selected as feature text items for various translation error detection.The experimental results show that the model constructed by the research institute has an error detection accuracy of over 95%for three types of English specialized vocabulary translations:general domain translation texts,specifi c domain translation texts,and colloquial translation texts,which is signifi cantly better than traditional methods.The above test results indicate this model has higher application reliability.

关 键 词：GLR算法 英语专用词汇 可分离性 翻译智能检错 模型构建 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] H319[自动化与计算机技术—计算机科学与技术]