The Ipe, Of Intelligent Translation Editor

The post-editing in machine translation is very significant. First, it is the main means to improve the quality of the target files of machine translation system. It provides many tools to help users to modify errors in the target files. Second, it is also the important basis for the accommodation of the machine translation system. By providing the feedback information to the learning mechanisms, it help the machine translation system to improve the quality of its knowledge bases.In this paper, we will propose a new model for post-editing — theintelligent post-editor IPE, and describe the design philosophy, overall architecture and principle algorithms of IPE in detail.As a system designed for post-editing specially, IPE has strong functions of editing. Furthermore, IPE also has some new features, when compared with other kinds of editors in the market. It provides users with convenient tools for them to post-edit. Using these tools, users can look up lexicons or translate a sentence easily. Besides, IPE also has the ability to learn from mistakes. It can record those modifications which the users made on the target files, and then acquire feedback information from those modifications. Using the feedback information, IPE can direct the learning mechanism to improve the knowledge bases of the machine translation system.With all these new features, IPE provides users with an effective and convenient post-editing environment. However, it still need some further development and maturation. so, in the end of this paper, we explore how the IPE can be further developed.