【作者】 邢富坤； 宋柔；

【Author】 Xing Fukun~1,Song Rou~2 1 PLA University of Foreign Languages,Luoyang 471003 2 Center for Language Information Processing,Beijing Language and Culture University,Beijing 100083

【机构】 解放军外国语学院； 北京语言大学语言信息处理研究所；

【摘要】 本文使用词性自动标注模型对影响英汉语词性标注的相关因素进行定量研究,进而探究词汇因素与语法因素各自对英汉语词性标注的影响,目的是为深入分析英汉语在词类问题上的差别,更好地构建汉语语料库提供参考依据。本文将词汇因素近似地形式化为词汇发射概率和词汇最大词性概率,语法因素近似地形式化为词性转移概率,并利用隐马尔科夫模型、马尔科夫模型和词汇最大概率模型进行自动标注实验。通过比较不同模型的标注准确率,发现语法因素对于英语词性标注的影响显著大于汉语,词汇因素对汉语词性标注的影响显著大于英语。在英汉语内部,英语的词的核心语义因素与句法因素对词性标注的影响基本相仿;而汉语的词的核心语义因素对词性标注的影响显著高于句法因素。更多还原

【Abstract】 This article makes a quantitive study on the basis of automatic PoS(Part of Speech) tagging model to explore the different roles of lexical and grammatical factors in English and Chinese PoS tagging.The main goal of this article is to make an in-depth study on the differences in the issue of word classes between English and Chinese and give some suggestions to the construction of Chinese corpus.The lexical factor is formalized as lexical emission probability and the grammatical factor is formalized as the PoS transition probability.Hidden Markov Model,Markov Model and Lexical Maximum Probability Model are used to do the PoS tagging experiment The experiment results show that the grammatical factor in English plays a more important role than that in Chinese.The lexical factor in Chinese has more significant effect on PoS tagging than that in English. In English lexical semantic factor and syntactic factor equally affect PoS tagging.In Chinese lexical semantic factor is more important than syntactic factor in PoS tagging.更多还原