中文信息学报基于词类串的汉语句子结构相似度计算方法.PDF

中　文 　信 　息 　学 　报 第 19 卷 第 1 期 　 JOURNAL OF CHINESE INFORMATION PROCESSING Vol 19 No1 ( ) 文章编号 :1003 - 0077 2005 01 - 0021 - 09 基于词类串的汉语句子结构相似度计算方法① 王荣波 ,池哲儒 (香港理工大学 电子及资讯工程系 多媒体信号处理中心 ,香港) 摘要 :句子相似度的衡量是基于实例机器翻译研究中最重要的一个内容 。对于基于实例的汉英机器翻译 研究 ,汉语句子相似度衡量的准确性 ,直接影响到最后翻译结果的输出。本文提出了一种汉语句子结构相似 性的计算方法 。该方法比较两个句子的词类信息串 ,进行最优匹配 ,得到一个结构相似性的值 。在小句子集 上的初步实验结果表明 ,该方法可行 ,有效 ,符合人的直观判断 。 关键词 :人工智能 ;机器翻译 ;基于实例机器翻译 ;汉英机器翻译 ;句子相似度衡量 ; 自然语言处理 中图分类号 :TP391 　　　文献标识码 :A A Similarity Measure Method of Chinese Sentence Structures WAN G Rongbo ,CHI Zheru (Center for Multimedia Signal Processing ,Department of Electronic and Information Engineering ,The Hong Kong Polytechnic University ,Hong Kong , China) ( ) Abstract :Examplebased machine translation EBMT is an important branch of machine translation that has been studied ex tensively for about twenty years. So far ,some progresses have been gained because of researchers hard work . Sentence simi larity measure certainly is one of the most important problems addressed in EBMT. For EBMT from Chinese to English ,the performance of similarity measure of Chinese sentences affects directly final translation result of an input sentence . In this pa per ,we proposed a similarity measure method of Chinese sentence structures for examplebased Chinese to English machine translation . In this method ,the algorithm performs the optimal matching between the word type sequences of two compared sen tences. The preliminary experimental results show that the measure method works well when it is tested on a small dataset . Key words : artificial intelligence ; mach