Translation agent

Abstract

When applying Language Grid for multilingual communication support, it is important to understand the effects of the machine translation service. To support the users in multilingual communication, it is recognized that machine translation service used is not a perfect transparent-channel to users. We present agent metaphor as a novel interactive way to promote the efficiency of communication in using machine translation service. It will reduce the communication breaks caused by translation errors. We propose to shift from the transparent-channel metaphor to the agent metaphor, which motives the interactions between the users and the machine translator. Following this paradigm shifting, the agent should encourage the dialog participants to collaborate, as their interactivity will be helpful in reducing the number of translation errors. We examined the translation issues raised by multilingual communication, and analyzed the impact of interactivity on the elimination of translation errors. We proposed an implementation of the agent protocol for interactivity promotion. We designed the architecture of our agent, analyzed the interaction process, and provided an example of preparing repair strategy. We conducted an English–Chinese communication task experiment on tangram arrangement. The experiment shows that compared to the transparent-channel metaphor, our agent metaphor reduces human communication effort by 21.6%.