改革开放后,新疆与国内以及国际交往越来越频繁。新疆少数民族人名汉字音译转写缺少统一标准,维吾尔人名汉字音译转写时,在户口上是一种写法,在身份证上另一种写法,在护照上更不一样的用字写法,机票、汇款单等又是一种写法。为解决这些问题,前人作了研究并取得了一定成果以及使用维吾尔人名汉语音译撰写系统来规范维吾尔人名汉语音译。但是这些成果正确率只有52%。利用规则和维吾尔人名特征相结合的方法,提高系统自动维吾尔人名汉语音译正确率30%。介绍该领域研究进展,通论维吾尔音节切分与维吾尔人名特征相结合方法。陈述基于字形的DOM音译框架解决的问题,介绍基于音节切分的维吾尔人名汉字音译转写的实现方法,给出了实验结果与分析。
After the reform and opening up, more frequent internal and international communication with Xinjiang occurs. The minority personal name transliteration to Chinese character is yet to have a unified standard, and various writings on residence registration, ID card, passport and even on ticket exist. To solve these problems, it has studied and obtained certain achievement using the system of Uyghur personal name transliteration to Chinese character. However, it only achieves 52% precision. Then using both rule and Uyghur personal name features combined approach, its precision increases over 30%. This paper introduces current research progress, then puts forward both Uyghur syllabification and Uyghur personal name features combined method, and states tackled issues of grapheme-based DOM transliteration frame. It gives the whole details of how this method runs on system of Uyghur personal name transliteration to Chinese characters ,and experi-mental result and analysis.