Research On Recognition And Translation Of Chinese-uyghur Time And Numeral And Quantifier

With the rapid development of the Internet Age,Named Entities and Machine Translation research has made break through progress.Now,In the Chinese-Uyghur machine translation appeared a lot of research work.In this paper proposed the translation methods based on the rules and templates of time,number,ambiguity-unambiguity quantification,and the translation method based on neural network with ambiguity quantification.This paper mainly studies the following three aspects:(1)According to the requirements of the Chinese-Uyghur translation and the Uyghur grammar structure.At the first,collecting and mining a large number of Chinese time expression rules and template library,the classification of digital collection(Ordinal Number?Approximate Number?Collective Number ? Fraction Number ? Percentage Number ? Multiple Number)constructing bilingual translation rules and template library respectively,to recognize Chinese time expression and templates.And find out its corresponding rules,output time expression or digital translation results.(2)For quantifiers are analyzed and classified,quantifier should be divided into ambiguity and unambiguous quantification,unambiguous quantifiers are divided into four kinds of circumstances(one-to-one correspondence,quantifier lost;digital one-to-one correspondence,quantifier one-to-one correspondence;number is missing,quantifier one-toone correspondence;digital and quantifiers are missing),the ambiguous quantifiers depending on the context of nouns,quantifier corresponding translation result is also different,through collecting the classified method of the quantifier phrase and increase the rule base.(3)Based on the Seq2 Seq model to realize the translation of neural machine translation with ambiguous quantifier phrases and improve the accuracy of Chinese-Uygur quantifiers.In order to improve Chinese machine translation system of time,Numbers,and performance of the quantifier phrase translation,bilingual corpus in this paper to extract the Chinese time,Numbers,and quantifier phrase,implements the translation method based on rules and neural network.In this paper.