语料对中文指代消解影响研究.docVIP

  语料对中文指代消解影响研究# 高俊伟，孔芳，朱巧明，李培峰** （苏州大学计算机科学与技术学院，江苏 苏州 215006） 5 10 15 20 25 30 摘要：指代是自然语言中一种常见的语言现象，对简化语言，减少冗余有很大的作用。指代 消解是用计算机找出这些指代现象的一个过程。近几年英文指代消解研究取得了很大的成 就，然而，中文指代消解研究目前还较少，一方面是由于中文自然语言处理的研究起步较晚， 相关的知识较少，另外一方面就是中文相关的语料库较少，目前已知的仅有 ACE2005, OntoNotes 等。为了探讨语料库对中文指代消解的影响，本文实现了一个基于机器学习方法 的中文名词短语指代消解平台和一个基于无监督聚类方法的中文名词短语指代消解平台，在 此平台的基础上从语料库的数量和质量两个方面来探讨语料对中文名词短语指代消解的影 响。 关键词：指代消解；名词短语；无监督；聚类；语料 中图分类号：TP391 Research on the effect of the corpus to the Chinese Noun Phrase Anaphora Resolution Gao Junwei, KONG Fang, ZHU Qiaoming, LI Peifeng (School of Computer Science Technology, Soochow University, Jiangsu, Suzhou 215006, China) Abstract: Coreference is a common phenomenon in natural language, it has a great effect that makes the natural language clear and explicit illusions. Coreference resolution is the process that finds these phenomenons by using the computer. A great deal of research has been done on this task in English and achieved a great achievement in recent years. However, much less work has been done in this area in Chinese. One problem is that the research of Chinese NLP is later than English, the other problem is that the lack of public corpus in the research of Chinese NLP, the public corpus of Chinese just have ACE2005, OntoNotes and so on. To discuss the effect of the corpus to the Chinese Noun Phrase Anaphora Resolution, we present a Chinese noun phrase coreference resolution system that based on machine learning approach and another system that based on unsupervised clustering approach. We discussed the effect of the corpus to the Chinese noun phrase coreference resolution based on the two platforms from the quantity and the quality of the corpus. Key words: Coreference resolution;Noun phrase;Unsupervised;Clustering;Corpus 35 0 引言 指代是自然语言中一种常见的语言现象，指在文本中用一个语言成分回指以前提及过的 某个语言单位，表现形式有别名，代词，专有名词等。这种指代现象对减少冗余，简化语言， 突出主题，使文本上下连贯通俗易懂都有很好的作用。随着人机对话，机器翻译，信息抽取 40 等研究的不断深入，指代消解的研究也越来越受到大家的关注，指代消解已成为自然语言处 理领域中的热点和难点之一。在语言学中，指代词也称为照应语，表示用于指向的语言单位， 被指向的