Evaluation Of Translation Memory Software Based On Context Theory

With the popularization of computer technology and network,computer-aided translation tools emerge one after another.At present,the mainstream computer-aided translation tool is translation memory software.The working principle of the translation memory software is that the user builds one or more translation memory databases by using the original text and the translation text.During the translation process,the system automatically searches for the same or similar paragraphs or sentences,and provides reference to the translation.It can make users avoid repeated translation,speed up the process of translation,thus reducing the amount of physical labor involved in translation projects.In the past,most of the studies on “translation memory” focused on technical improvement.In the past two decades,translation memory has entered the field of foreign language research and has been applied in translation teaching.However,the evaluation of translation memory software is still a weak step.Taking Trados and snowman as examples,this paper attempts to propose an evaluation model of translation memory software.(1)Theoretical method: the context theory of Melby & Foster is adopted.It includes five elements--co-text,chrono-text,bi-text,re-text and non-text,which can depict the functional modules of translation and memory software.(2)Functional module analysis: based on the above context theory,the functional modules of translation memory software(mainly including translation memory database,term database,alignment tool,translation work interface and matching mechanism)are analyzed and evaluated.On the one hand,it focuses on revealing the characteristics of the functional modules of Trados and Snowman,especially what kind of elements of context theory they embody respectively.On the other hand,from the perspective of the relationship between context theory and function module,the contextual effectiveness of Trados and Snowman is evaluated and compared.Based onthat,the result that which software is better will be clear.(3)Experimental design: Taking the Constitution of the People's Republic of China as a material sample.Load it into the two software of Trados and Snowman respectively for investigating the difference of the accuracy and efficiency of translation.On this basis,the relationship between contextual effectiveness and the accuracy and efficiency of the translation memory software is further studied.The feasibility of the evaluation model based on the context theory will be investigated.The results are as follows:(1)after comparing the functional analysis of translation memory software and the elements of context theory,both Trados and Snowman embody the four elements of context theory--co-text,chrono-text,bi-text and rel-text;(2)each functional module of the translation memory software corresponds to a number of elements of context theory,and each context element also corresponds to a number of functional module,so there is a many-to-many relationship between them;(3)the contextual effectiveness embodied by Trados is better than that of snowman;(4)the contextual effectiveness is proportional to the translation accuracy and efficiency of the translation memory software,so the evaluation model based on the context theory is feasible;(5)from the perspective of contextual effectiveness,accuracy and efficiency of translation,Trados is better than Snowman.Through the above-mentioned research work,the explanatory power of the model is demonstrated to some extent.This study provides good suggestions for the relevant technicians to further improve the translation memory technology.In addition,since this study combines linguistic theory with translation technology more closely,it is helpful to carry out the teaching of translation memory in the future.