时间表达式识别是进行时间表达式归一化的基础,其识别结果的好坏直接影响归一化的效果。本文提出一种基于依存分析和错误驱动识别中文时间表达式的新方法。首先以时间触发词为切入点,据依存关系递归地识别时间表达式,大大地提高了识别效果;然后,采用错误驱动学习来进一步增强识别效果,根据错误识别结果和人工标注的差异自动地获取和改进规则,使系统的性能又提高了近3.5%。最终在封闭测试集和开放测试集上,F1值达到了76.38%和76.57%。
Abstract
Recognizing time expressions is the foundation of its normalization, and its performance directly influences the robustness of the normalization. This paper proposes a new method for recognizing the extents of the time expressions based on dependency parsing and error-driven learning, which begins with time trigger word (namely, the syntactic head of dependency relation), uses Chinese dependency parsing to recognize the extents of the time expressions, Subsequently, we use the transformation-based error-driven learning to improve the performance., which can automatically acquire and modify the rules and get 3.5% increase after applying the learned rules. Finally, F1 = 76.38% and F1 =76.57% results are obtained on the closed and the open test set respectively.
关键词
计算机应用 /
中文信息处理 /
时间表达式识别 /
触发词 /
依存分析 /
错误驱动学习
{{custom_keyword}} /
Key words
computer application /
Chinese information processing /
time expression recognition /
trigger word /
dependency parsing /
error-driven learning
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Mingli Wu, Wenjie Li, Qin Lu, Baoli Li. CTEMP: A Chinese Temporal Parser for Extracting and Normalizing Temporal Information[A]. IJCNLP 2005 [C]. 694-706.
[2] Yang Ye, Victoria Li Fossum, and Steven Abney. Latent features in automatic tense translation between chinese and english[A]. In: Proceedings of the Fifth SIGHAN Workshop on Chinese Language Processing [C]. Sydney, Australia: July 2006. 48-55.
[3] SemEval-2007[EB/OL]. http://nlp.cs.swarthmore.edu/semeval/index.shtml.
[4] Vazov N. A System for Extraction of Temporal Expressions from French Texts based on Syntactic and Semantic Constraints[A]. In: Proceedings of the ACL Workshop on Temporal and Spatial Information Processing (2001) [C]. 96-103.
[5] Wilson, G., Mani, I., Sundheim, B., and Ferro, L. 2001. A multilingual approach to annotating and extracting temporal information[A]. In: Proceedings of the Workshop for Temporal and Spatial Information Processing EACL-ACL 2001 [C]. Toulouse, France: July, 2001.
[6] SETZER, A. 2001. Temporal information in newswire articles: An annotation scheme and corpus study[EB/OL]. Ph.D.thesis, Univ. of Sheffield.
[7] Mani, I. 2004. Recent Developments in Temporal Information Extraction[A]. In: NICOLOV, N., AND MITKOV, K., Proceedings of the Conference on Recent Advances In Natural Language Processing [C]. John Benjamins.
[8] ACE2007 evaluation plan[EB/OL]. http://projects.ldc.upenn.edu/ace/intro.html 2006-11-6.
[9] Jang, S.B., Baldwin, J. and Mani, I.: Automatic TIMEX2 Tagging of Korean News[J]. ACM Transactions on Asian Language Information processing 2004, 3(1): 51-65.
[10] Estela, S., Martinez-Barco, Patricio, and Munoz, R.: Recognizing and Tagging Temporal Expressions in Spanish[A]. Workshop on Annotation Standards for Temporal Information in Natural Language, LREC 2002 [C].
[11] Hacioglu, Kadri, Chen, Ying and Douglas , Benjamin, Automatic Time Expression Labeling for English and Chinese Text[A]. To appear in Proceedings of CICLing-2005 [C]. Mexico City-Mexico: Feb. 13-19, 2005.
[12] Ahn, D., Adafre, S. F., and Rijke, M. de.: Towards Task-Based Temporal Extraction and Recognition[A]. In: Proceedings Dagstuhl Workshop on Annotating, Extracting, and Reasoning about Time and Events (2005) [C].
[13] Allen J.F., Towards a General Theory of Action and Time[J]. Artificial Intelligence, 1984, 23: 123-154.
[14] Brill , Eric. Transformation-based error-driven learning and natural language processing: A case study in part of speech tagging[J]. Computational Linguistics,1995, 21(4):543-565.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金资助项目(60575042)
{{custom_fund}}
PDF(265 KB)
