Example-based Chinese-English MT

Chinese sentences are translated into English sentences using example-based machine translation. The main process of example-based machine translation is divided into three phases. First, find the most similar examples as the input sentence. Then, recombine the translation of the input sentence according to most similar example and bilingual dictionary. Lastly, produce the translation of the input sentence. The main resources are bilingual dictionary, thesaurus, the standard template system and bilingual sentence aligned corpora. The bilingual dictionary, thesaurus, and the standard template system are used to calculate the similarities between two words, two chunks, and two sentences. At the same time, bilingual dictionary and bilingual sentence aligned corpora are used to adjust and produce the translation. Future improvement is given.