计算机与现代化 ›› 2012, Vol. 203 ›› Issue (7): 198-201.doi: 10.3969/j.issn.1006-2475.2012.07.055

• 应用与开发 • 上一篇    下一篇

面向文本的事件信息抽取方法的研究

刘敬培1,李江1,季文平1,潘鹏辉2   

  1. 1.渭南供电局通信中心,陕西渭南714000;2.西安微电子技术研究所,陕西西安710054
  • 收稿日期:2012-02-23 修回日期:1900-01-01 出版日期:2012-08-10 发布日期:2012-08-10

Research on Text-oriented Event Information Extraction Method

LIU Jing-pei 1, LI Jiang1, JI Wen-ping1, PAN Peng-hui2   

  1. 1.Information and Communication Centre, Weinan Power Supply Bureau, Weinan 714000, China; 2.Xi’an Institute of Microelectronics Technology, Xi’an 710054, China
  • Received:2012-02-23 Revised:1900-01-01 Online:2012-08-10 Published:2012-08-10

摘要: 研究面向文本的事件信息抽取工作,建立一个事件信息抽取系统。该系统首先过滤包含关键字的原始语料;然后采用层次聚类(Hierarchical,HCL)和最长公共子序列算法相结合的方法抽取事件信息,得到最初的模式;最后通过是否包含关键字进行模式获取,进而提取信息,最终得到事件要素。

关键词: 信息抽取, 事件信息抽取, 模式

Abstract: This paper studies the extraction of text-oriented event information, establishes an event information extraction system. This system filters original corpus contain keywords firstly, obtains initial patterns using the method combined with hierarchical clustering (HCL) and the algorithm of longest common subsequence secondly, and then obtains final patterns contain keywords. Using these patterns information can be extracted and elements of event can be obtained finally.

Key words: information extraction, event information extraction, pattern

中图分类号: