计算机工程 ›› 2007, Vol. 33 ›› Issue (21): 207-209.doi: 10.3969/j.issn.1000-3428.2007.21.074

• 人工智能及识别技术 • 上一篇    下一篇

基于本体关系匹配的信息抽取

何召卫,陈俊亮   

  1. (北京邮电大学网络与交换技术国家重点实验室,北京100876)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-05 发布日期:2007-11-05

Information Extraction Based on Ontology Entity Relation Match

HE Zhao-wei, CHEN Jun-liang   

  1. (State Key Laboratory of Networking & Switching Technology, Beijing University of Posts & Telecommunications, Beijing 100876)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-05 Published:2007-11-05

摘要: 目前,稳定可靠的信息抽取是一个有待解决的问题,该文提出了基于本体关系匹配信息抽取算法,应用语义Web把信息抽取目标文档描述为特殊的本体格式,采用机器学习理论对本体进行分析和处理。测试数据集的实验结果显示,本体关系集匹配算法优于其他4种信息抽取算法。

关键词: 信息抽取, 语义Web, 机器学习

Abstract: At present, how to extract information reliably is still a difficult, unsolved problem. Based on ontology entity relation match, this paper presents an information extraction new algorithm based on ontology entity relation match, which presents the object document to special ontology of semantic Web, analyzes and processes the ontology by machine learning principle. Experimental results show that relation match algorithm based on ontology have an advantage over other four approaches.

Key words: information extraction, semantic Web, machine learning

中图分类号: