Research And Implementation Of An AI-Assisted Analysis System For English Translation Teaching

With the development of artificial intelligence technology,the level of machine translation has been greatly improved,which poses new challenges to the cultivation of translation talents.In the future,the teaching of translation will focus on examining students’ ability to review machine translation on the basis of examining students’ translation ability.Existing online education systems are mainly divided into general systems that realize basic teaching functions and customized systems for specific fields.The functions of the general system are relatively basic and cannot effectively cover specific sub-fields;the customized system is currently mainly for composition,grammar and other fields,and there are few translation teaching platforms,which do not include systems for new translation teaching scenarios.Therefore,it is very important for future translation teaching to form an education system that is customized for new translation teaching scenarios and has certain intelligent analysis and processing capabilities.This paper aims to implement an English translation teaching assistant system that supports intelligent analysis.In the part of intelligent analysis and processing,the corresponding detection methods are firstly proposed for several mistakes that students often make in translation:article error detection based on semantic features and neural network model,missing translation error detection based on information content and synonym set,and spelling error detection based on the improvement of the Aspell;Secondly,analyze from the perspective of the whole error,generate association rules between errors,divide the error community to which the error belongs.The combination of the two draws the connection and rules between errors;after that,the students are divided into clusters by fusion,and typical errors corresponding to each category are generated.Finally,the article community is constructed and the corresponding core knowledge points are summarized,the collaborative filtering homework recommendation is carried out according to the students’ category,and the results are further screened according to the error situation of the target students.This paper also introduces the design and implementation of the system in detail.According to the demand analysis of the system according to the new translation teaching scene,the overall design structure of the system is given.According to the functional design,the modules are divided into article,homework,error feedback and induction,class,account,and intelligent analysis and processing.Finally,the system is functionally tested and verified,and the actual online usage of the system is introduced.