计算机科学 ›› 2015, Vol. 42 ›› Issue (2): 253-255.doi: 10.11896/j.issn.1002-137X.2015.02.052

• 人工智能 • 上一篇    下一篇

半监督中文事件抽取中的模板过滤和转换方法

徐霞,李培峰,朱巧明   

  1. 苏州大学计算机科学与技术学院 苏州215006,江苏省计算机信息处理技术重点实验室 苏州215006,江苏省计算机信息处理技术重点实验室 苏州215006
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家自然科学基金(61272260),江苏省自然基金(BK2011282),江苏省高校自然科学重大基础研究项目(11KIJ520003)资助

Pattern Filtering and Conversion Methods for Semi-supervised Chinese Event Extraction

XU Xia, LI Pei-feng and ZHU Qiao-ming   

  • Online:2018-11-14 Published:2018-11-14

摘要: 事件模板是指导事件抽取工作的依据,半监督方法下模板的准确性显得尤为重要。目前,基于双视图的“触发词-论元”模板的中文信息事件抽取系统不能有效地解决触发词一词多义的现象和模板稀疏现象。提出了一种借助论元进行触发词语义消歧的方法,并利用该方法进行模板过滤以消除无效模板的影响。另外,针对几种特殊的中文句型,根据句法结构提出了模板转换规则,从而提高了模板的适用性。在ACE2005中文语料上的测试表明,该方法可有效地提高半监督中文信息事件抽取系统的性能。

关键词: 事件抽取,模板过滤,模板转换

Abstract: The accuracy of event patterns is very important in semi-supervised event extraction.Currently,semi-supervised Chinese event extraction system based on the pairwise pattern (e.g.,Trigger-Argument) suffers much from the issues of polysemy of triggers and sparse patterns.This paper put forward a argument-based mechanism to solve trigger sense disambiguation,and then applied it to pattern filtering to eliminate invalid patterns.In addition,for several special Chinese sentence structures,this paper proposed a pattern conversion method based on syntactic structure to enhance the applicability of the pattern.The experimental results on the ACE 2005 Chinese data show that our methods can effectively improve performance of semi-supervised Chinese event extraction system.

Key words: Event extraction,Pattern filtering,Pattern conversion

[1] Riloff E.Automatically Generating Extraction Patterns fromUntagged Text[C]∥Proceedings of the Thirteenth National Conference on Artificial Intelligence.1996:1044-1049
[2] Yangarber R,Grishman R,Tapanainen P,et al.Automatic Acquisition of Domain Knowledge for Information Extraction[C]∥Proceedings of the 18th Conference on Computational linguistics.2000:940-946
[3] Yangarber R.Counter-Training in Discovery of Semantic Pat-terns[C]∥Proceedings of ACL 2003.2003:343-350
[4] Huang Rui-hong,Riloff E.Bootstrapped Training of Event Extraction Classifiers[C]∥Proceedings of EACL 2012.2012:286-295
[5] Phillips W,Riloff E.Exploiting Role-Identifying Nouns and Expressions for Information Extraction[C]∥Proceedings of RANLP 2007.2007:468-473
[6] Stevenson M,Greenwood M.A Semantic Approach to IE Pat-tern Induction[C]∥Proceedings of ACL 2005.2005:379-386
[7] Liao Sha-sha,Grishman R.Filtered Ranking for Bootstrapping in Event Extraction[C]∥Proceedings of COLING 2010.2010:680-688
[8] Liao Sha-sha,Grishman R.Can Document Selection Help Semi-supervised Learning? A Case Study on Event Extraction[C]∥Proceedings of ACL 2011.2011:260-265
[9] Chen Zheng,Ji Heng.Can One Language Bootstrap the Other:A Case Study on Event Extraction[C]∥Proceedings of HLT-NAACL 2009 Workshop on Semi-supervised Learning for Natural Language Processing.2009:66-74
[10] Chambers N,Jurafsky D.Unsupervised Learning of NarrativeEvent Chains[C]∥Proceedings of ACL-HLT 2008.2008:787-797
[11] Chambers N,Jurafsky D.Unsupervised Learning of NarrativeSchemas and Their Participants[C]∥Proceedings of ACL 2009.2009:602-610
[12] Balasubramanian N,Soderland S,Mausam,et al.Generating Coherent Event Schemas at Scale[C]∥Proceedings of the 2013 Conference on Empirical Methods in Natural Language Proces-sing.2013:1721-1731
[13] Kiyoshi S,Satoshi S,Ralph G.Automatic Pattern Acquisitionfor Japanese Information Extraction[C]∥Proceedings of HLT 2001.2001:1-7
[14] Kiyoshi S,Satoshi S,Ralph G.An Improved Extraction Pattern Representation Model for Automatic IE Pattern Acquisition[C]∥Proceedings of ACL 2003.2003:224-231
[15] Liu Ting,Strzalkowski T.Bootstrapping Events and Relationsfrom Text[C]∥Proceedings of EACL 2012.2012:296-305

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!