计算机工程 ›› 2009, Vol. 35 ›› Issue (15): 165-167.doi: 10.3969/j.issn.1000-3428.2009.15.057

• 人工智能及识别技术 • 上一篇    下一篇

基于树核函数的代词指代消解

王海东1,2,谭魏璇1,2,李艳翠1,2,周国栋1,2   

  1. (1. 苏州大学计算机科学与技术学院,苏州 215006; 2. 江苏省计算机信息处理技术重点实验室,苏州 215006)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-08-05 发布日期:2009-08-05

Tree Kernel Function-based Pronoun Coreference Resolution

WANG Hai-dong1,2, TAN Wei-xuan1,2, LI Yan-cui1,2, ZHOU Guo-dong1,2   

  1. (1. School of Computer Science & Technology, Soochow University, Suzhou 215006; 2. Jiangsu Province Key Lab for Computer Information Processing Technology, Suzhou 215006)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-08-05 Published:2009-08-05

摘要: 提出一种基于树核的英文代词消解方法。针对结构化信息在指代消解中的重要作用,使用SVM提供的卷积树核函数自动获取句法结构信息,将句法树作为一个特征与其他基本特征结合。通过应用不同的剪枝策略,考虑不同句法树对系统的影响,在原有的句法树上扩充一些语义节点。在ACE2004 NWIRE基准数据上进行实验的结果证明,该方法对代词的消解起到明显的作用,综合值f提高了11.9%。

关键词: 指代消解, 句法结构, 树核函数, 修剪策略

Abstract: This paper proposes a tree kernel-based approach to anaphora resolution of pronoun. On the basis of structured information automatically captured by convolve kernel of SVM, it integrates syntax tree as a feature with other base features. Different pruning strategies are applied to eliminate the impact of syntax trees to the results. Evaluation on the ACE2004 NWIRE benchmark corpus shows that tree kernel can improve the f performance by 11.9%. Based on the system, it combines with semantic role feature and verb-driving feature which are acquired from ASSERT.

Key words: coreference resolution, syntax structure, tree kernel function, pruning strategy

中图分类号: