基于网络搜索的英汉人名翻译

发布时间：2018-04-03 02:10

  本文选题：人名翻译　切入点：音译相似度　出处：《中文信息学报》2017年02期

【摘要】：该文利用搜索引擎从网络中挖掘英语人名的中文翻译。该方法综合利用翻译辅助词、英中人名共现规则、音译相似度和翻译概率。首先,利用搜索引擎从互联网上搜索英文人名的中文翻译候选。把汉语人名标注结果、翻译辅助词、英中人名共现规则和英文人名的发音音节长度结合起来提取翻译候选词。翻译辅助词有助于搜索与英文人名更相关的信息,英中人名共现规则和发音音节长度进一步缩小英文人名的翻译范围,使得英文人名的翻译搜索符合人名共现规律和发音规律。然后,根据音译相似度和翻译概率对候选词进行排序。人名翻译的绝大部分是根据发音翻译过来的,音译相似度是帮助判断两个词在发音上的相似性。翻译概率从统计上判断两个词互为翻译的可能性。实验结果表明,翻译辅助词、规则、音译相似度和翻译概率都有助于提高人名翻译的正确率。
[Abstract]:This paper uses search engine to excavate the Chinese translation of English names from the network.The method combines translation auxiliary words, English and Chinese cooccurrence rules, transliteration similarity and translation probability.First, search engines are used to search the Chinese translation candidates for English names.Combining the results of Chinese personal name tagging, translation auxiliary words, co-occurrence rules of English and Chinese names and the pronunciation syllable length of English names, the candidate words for translation are extracted.Translation auxiliaries are helpful to search for information related to English names. The rules of co-occurrence of English and Chinese names and the length of pronunciation syllables further reduce the translation scope of English names, so that the translation search of English names conforms to the law of cooccurrence of names and the rule of pronunciation.Then, the candidate words are sorted according to transliteration similarity and translation probability.The transliteration similarity is to help judge the pronunciation similarity of two words.The probability of translation determines statistically the possibility of two words being translated into each other.The experimental results show that translation auxiliary words, rules, transliteration similarity and translation probability can improve the accuracy of human name translation.
【作者单位】： 清华大学中文系;
【基金】：国家自然科学基金(61171114) 北京市社科基金(16YYB021) 清华大学人文社科振兴基金(20145081042)
【分类号】：TP391.1
，

本文编号：1703145