基于事件模型的省略恢复研究  

Study on Ellipsis Recovering Based on Event Model

作　　者：许威[1] 赵克[1] 李亚涛[1] 亿珍珍[1] 

机构地区：[1]西安电子科技大学机电工程学院,陕西西安710071

出　　处：《郑州大学学报（理学版）》2011年第2期32-37,共6页Journal of Zhengzhou University:Natural Science Edition

摘　　要：研究了目前基于事件模型的汉语省略恢复,重点分析了自然语言的事件层次,认为事件是一个具有一定句法结构、语义功能并集成了相关的语境信息的语言单位.建立了用于表示事件内涵、外延及其二者关系的事件模型和动词语义知识的事件描述,从事件结构的语法到语义层面进行映射,对缺省的语义成分进行判定,构建一个省略恢复模型,并运用在自然语言理解领域中,结果表明,在汉语正式体省略恢复中具有一定优越性.Chinese ellipsis recovering based on event model was studied. The event level of natural language was analyzed. The event was considered as a language unit which had certain syntax structure and semantic function, and was integrated the context information. An event model adapted for representing the event connotation, the extension and the relation between the connotation and the extension was presented. Through the mapping from syntax level to semantic level, an ellipsis recovering model was proposed which was applied to the domain natural language understanding. Simulation results showed the advantages of the proposed approach in Chinese formal ellipsis recovering.

关 键 词：省略 事件 事件模型 自然语言理解 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]