Translation Technologies of Portuguese to Chinese Machine Translation System: PCT Assistente

This paper presents the translation technologies of Portuguese to Chinese machine translation system, PCT Assistente a hybrid translation model integrated with example-based and rule-based translation paradigms to counterbalance the intrinsic weakness of these approaches by combining the strong features of another. In the example-based model, translation examples are annotated under the schema of Translation Corresponding Tree (TCT) structure that acted as the basic knowledge for performing the translation task. While in the rule-based model, Constraint Synchronous Grammar (CSG) is adopted as the language formalism to analyze the syntactic structure of an input sentence and produce the corresponding translation in target language during the translation process. In this paper, the translation paradigms based on Translation Corresponding Tree and Constraint Synchronous Grammar are discussed. The linkage of two translation modules by combining the strengths of different approaches to improve the quality of the translation system is studied, and finally, the architecture of network based PCT Assistente (machine translation) system is described.