人物简历汉维机器翻译系统的设计与实现

龙源期刊网 https://www.sodocs.net/doc/b216823852.html,

人物简历汉维机器翻译系统的设计与实现

作者：王路路斯拉吉艾合麦提·如则麦麦提艾山·吾买尔吐尔根·依布拉音买合木提·买买提卡哈尔江·阿比的热西提

来源：《现代电子技术》2018年第24期

关键词：人物简历; 汉维机器翻译; 模板库; 命名实体; 正则表达式; 句子结构

中图分类号： TN912.3?34; TP391 ; ; ; ; ; ; ; ; 文献标识码： A ; ; ; ; ; ; ; ; ;文章编号：

1004?373X（2018）24?0101?05

Design and implementation of Chinese?Uyghur machine translation

system for personage resumes

WANG Lulu1，2， Silajiaihemaiti Ruzemaimaiti1，2， Aishan Wumaier1，2， Tuergen Yibulayin1，2，

Maihemuti Maimaiti1，2， Kahaerjiang Abiderexiti1，2

（1. School of Information Science and Engineering， Xinjiang University， Urumqi 830046，China;

2. Xinjiang Laboratory of Multi?Language Information Technology， Urumqi 830046，China）

Abstract： A Chinese?Uyghur machine translation system based on the combination of templates and dictionaries is designed and implemented for personage resumes to resolve the problems of high cost and low efficiency of translating the Chinese personage resume into the Uyghur version. The named entities are generalized， and the Chinese resume templates are automatically created by analyzing the sentence structure characteristics of Chinese resumes. A Chinese?Uyghur template library is established by using regular expressions. The named entities are translated by combining with the method based on rules and dictionaries， so as to obtain the Uyghur personage resumes. The experimental results show that the system uses limited templates and dictionaries in the monolingual situation， its BLUE value can reach 0.38， and in comparison with the machine translation systems that require large?scale bilingual corpora training， the system has a better practical application value.

Keywords： personage resume; Chinese?Uyghur machine translation; template library; named entity; regular expression; sentence structure