跳到主要导航 跳到搜索 跳到主要内容

Entity relationship extraction based on potential relationship pattern

  • Chen Chen*
  • , Huilin Liu
  • , Guoren Wang
  • , Linlin Ding
  • , Lili Yu
  • *此作品的通讯作者
  • Northeastern University China
  • Neusoft Corporation

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

The keep rising of web information ensures the development of entity focused information retrieval system. However, the problem of mining the relationships effectively between entities has not been well resolved. For the entity relationship extraction (RE) problem, this paper firstly establishes the basic pattern trees which can present the overall relation structures and then designs a similarity function according to which we can judge which pattern the sentence containing two entities belongs to. Knowing the matched pattern, we can discovery the relationship easily. By a large number of experiments on real data, the proposed methods are proved running accurately and efficiently.

源语言英语
主期刊名Web Information Systems and Mining - International Conference, WISM 2010, Proceedings
370-377
页数8
版本M4D
DOI
出版状态已出版 - 2010
已对外发布
活动2010 International Conference on Web Information Systems and Mining, WISM 2010 - Sanya, 中国
期限: 23 10月 201024 10月 2010

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
编号M4D
6318 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议2010 International Conference on Web Information Systems and Mining, WISM 2010
国家/地区中国
Sanya
时期23/10/1024/10/10

指纹

探究 'Entity relationship extraction based on potential relationship pattern' 的科研主题。它们共同构成独一无二的指纹。

引用此