నైరూప్య

Research on machine translation based on key technologies of bilingual corpus

Di Lu

With the development of the technology of statistical natural language processing, the role of parallel corpus in statistical machine translation and cross-language retrieval cannot be ignored. In this paper, we examines the translation equivalent pairs could be extracted from parallel corpus. An iterative algorithm based on degree of word association is proposed to identify the multiword units for Chinese and English. Then a hypothesis testing approach is used to extract the Chinese English Translation Equivalent Pairs. We present a tree-tree model by mapping between the syntactic tree and the ITG tree, the model limits the reordering of the phrases in the global scope. While in the local scope, the tree-tree model takes the TTG-based local reordering model as one feature, in which the reordering probability of two blocks is decomposed into the product of the reordering probabilities of the child blocks respectively. So the model is able to estimate the reordering of two blocks with arbitrary lengths.