Information Technology In Education-oriented Chinese-Mongolian Machine Translation Research

Machine translation is a natural language processing is an important application areas. Machine translation to achieve access to information, information exchange, information sharing, and so has important application value.The statistical machine translation methods mainly include word-based translation model、 phrase-based translation model and syntax-based translation model. The phrase-based translation model is improved a lot than word-based translation model on the quality of translation, and it also has the advantages of having simple model, training fast and decoding efficiently, compared with syntax-based translation model. Therefore, phrase-based statistical machine translation method has gradually become the hot topic in the field of statistical machine translation.This thesis discusses the relative theories of statistical machine translation at first, and then we use the method of phrase-based statistical machine translation and international existing open source software to build a Mongolian-Chinese statistical machine translation system based on phrase, combining the status of Mongolian information processing technology. This system will automatically translate Mongolian text into Chinese text.Secondly, we established a development set and test set,Then obtained a certain amount of Mongolian and Chinese parallel corpus through the two approaches of collecting and establishing.On.Finally, according to the above problems, to improve the performance of Mongolian Chinese machine translation.