基于音节双向最大化匹配的英汉名称音译

English-Chinese Name Transliteration with Bi-Directional Syllable-Based Maximum Matching

目前的众多音译系统中,比较流行的主要有两种方法:基于短语统计的机器音译法和将翻译对象作为一种标签序列的条件随机域方法。相比于流行方法,研究一种更加简单直观、更具语言性和实用性、通用性的英汉名称音译方法 (音节双向最大化匹配),并建立基于此方法的双语词库和音译系统意义深远。

There are two particularly popular approaches to transliteration which are phrase-based statistical machine transliteration and the Conditional Random Fields in the present study. Compared with the popular research method,the study of a more simple and more intuitive universal English name transliteration method（syllable bidirectional maximum matching）,which is of practical language,and a bilingual lexicon and transliteration system,which are based on this method have far-reaching significance.