Statistical machine translation proposal for Uzbek to English

Science and Education

Alisher Shakirovich Ismailov¹  Gulshoda Shamsiyeva²  Nilufar Abdurakhmonova³ 

[1] Andijan Machine Building Institute;National University of Uzbekistan;Tashkent State University of Uzbek Language and Literature named after Alisher Navoi;

关键词: machine translation;    natural language;    statistical machine translation;    corpora;   

DOI  :  

来源: DOAJ

【 摘 要 】

The machine translation means is a translating one natural language to another natural language automatically [1]. The machine translation is one of the major and the most active areas in natural language processing. The last decade have seen the rise of the use of statistical approaches to the machine translation. The statistical machine translation approaches learn translation parameters automatically from alignment text rather than relying on rule-based approaches. There has been quite extensive work in statistical machine translation area for some language pairs. However, there are very limited research sources available for the Uzbek to English language pair [2]. In this paper, we propose statistical machine translation algorithm for Uzbek to English. The developing English to Uzbek statistical machine translation algorithm is an interesting obstacle from a number of perspectives. The most important challenge is that English and Uzbek are typologically distant languages. The English language has very limited morphology and Uzbek is an agglutinative language with a very rich and productive derivational and inflectional morphology. The Uzbek word structures that can correspond to complete phrases of several words in English when translated. In this paper, propose that will achieve Uzbek to English statistical machine translation algorithm using phrase-base model. Moreover, in order to achieve statistical machine translation we need to develop English-Uzbek corpora. In this paper, we present briefly about English-Uzbek corpora development.

【 授权许可】

Unknown