A Neural Network Mongolian-Chinese Machine Translation Method

Abstract

Provided is a neural network Mongol-Chinese machine translation method. The method includes the steps of firstly, conducting standardized processing on large-scale Mongol-Chinese bilingual corpora, and on this basis, constructing out a bilingual dictionary; then, conducting modelling, and finally, based on a constructed model, achieving machine translation. The method is characterized in that modeling includes encoder modeling, decoder modeling and attention layer modeling needed by Mongol-Chinese bilingual word alignment, and therefore, aiming at Mongol terms with specific word characteristics, corresponding processing is conducted to achieve the best translation effect and the smallest semantic confusion. According to the method, firstly, Mongol lexical information is contained in the translation model, and accurate depiction is conducted on Mongol encoding by using an encoder network; secondly, an algorithm ensures the quality of Mongol-Chinese translation; finally, by applying a neural network, Mongol-Chinese bilingual translation is disintegrated into Mongol encoding and Chinese decoding, a neural network algorithm with high extendibility is proposed, and the version fluency and translation quality of Mongol-Chinese machine translation are effectively improved.

Description

technical field [0001] The invention belongs to the technical field of machine translation, and in particular relates to a neural network Mongolian-Chinese machine translation method. Background technique [0002] With the vigorous development of a series of network services including information technology, natural language processing, especially machine translation, plays a vital role in the development of the Internet. Many large search companies and service centers such as Google and Baidu have conducted large-scale research on machine translation, making unremitting efforts to obtain high-quality translations of machine translation. [0003] However, with the continuous efforts of scientists and linguists for decades, problems that cannot be ignored have also been exposed in the process of machine translation development - ambiguous word processing, unregistered word processing, and coding confusion caused by differences in bilingual structure. The translation quality ...

AI Technical Summary

Problems solved by technology

[0007] In order to overcome the above-mentioned shortcomings such as missing translations, mistranslations, and unregistered word processing in the translation process, the purpose of the present invention is to provide a neural network Mongolian-Chinese machine translation method, aiming at the scarcity of data in the small corpus and the small size of the dictionary. In order to reduce the complexity of the system and ensure the quality of translation services for users under the condition of reducing the complexity of the system and visualizing the system structure for users, we can improve the Mongolian-Chinese machine translation system and achieve the goal of better translation

Images