本文提出了一种结合SVM学习和产生式模型的依存分析方法。该方法用产生式模型的分析错误对SVM分类器进行训练。为进一步提高分析精度,采用扩大寻优范围的动态规划算法对产生式模型的分析结果进行错误估计,同时引入范围参数,使得寻优范围可以根据实际情况进行调整。本方法在不牺牲分类性能的前提下,有效减少了训练SVM分类器所依赖的支撑向量数。在对哈工大中文树库语料上的对比测试结果表明,该方法的依存分析精度达到86.4%,具有很强的依存分析能力。
Abstract
In this paper, we propose a SVM-combined generative statistical model for Chinese dependency analysis that trains SVM classifier using erroneous results generated by generative statistical model. To further improve the precision of dependency analysis, two measures were taken, first, dynamic programming algorithm that extends the range of finding the best local solution was used to estimate the error rate of generative model; second, a ranging factor was introduced to make the solutions adaptive on the practical situation. All those efforts make it possible for the new method to largely decrease the number of negative support vectors without sacrificing classification ability in training. Comparative experiments on Hit Chinese Treebank corpus show that the new method shows better performance than current Chinese dependency methods, with precision reaching to 86.4%.
关键词
计算机应用 /
中文信息处理 /
中文依存分析 /
产生式概率模型 /
SVM学习 /
SMO /
动态规划算法
{{custom_keyword}} /
Key words
computer application /
chinese information processing /
chinese dependency analysis /
generative statistical model /
SVM study /
SMO (Sequential Minimal Opeimization) /
dynamic programming algorithm.
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Gaifman, H. Dependency systems and phrase-structure systems[J]. Information and Control, 1965,8: 304-337.
[2] Jrvinen, T. and Tapanainen, P. Towards an implementable dependency grammar[A]. Kahane, S. and Polgu`ere, A. (eds), In: Proceedings of the Workshop on Processing of Dependency-Based Grammars [C]. 1998. 1-10.
[3] Nivre, J. An efficient algorithm for projective dependency parsing[A]. Van Noord, G. (ed.), In: Proceedings of the 8th International Workshop on Parsing Technologies (IWPT) [C]. 2003. 149-160.
[4] Eisner, J. M. Three new probabilistic models for dependency parsing[A]. In: An exploration. Proceedings of the 16th International Conference on Computational Linguistics (COLING) [C]. 1996, 340-345.
[5] Eisner, J. M. Bilexical grammars and their cubic-time parsing algorithms[A].In: Bunt, H. and Nijholt, A. (eds), Advances in Probabilistic and Other Parsing Technologies [C]. Kluwer, 2000, 29-62.
[6] Samuelsson,C. A statistical theory of dependency syntax [A]. In: Proceedings of the 18th International Conference on Computational Linguistics (COLING) [C]. 2000.
[7] Yamada, H. and Matsumoto, Y. Statistical dependency analysis with support vector machines[A]. Van Noord, G. (ed.), In: Proceedings of the 8th International Workshop on Parsing Technologies (IWPT) [C]. 2003, 195-206.
[8] Kudo, T. and Matsumoto, Y. Japanese dependency analysis using cascaded chunking[A]. In: Proceedings of the Sixth Workshop on Computational Language Learning (CoNLL) [C]. 2002, 63-69.
[9] XU Yun, Zhang Feng. Using SVM to Construct a Chinese dependency parser[J]. Journal of Zhejiang University Science A, 2006, 7(2): 199-203.
[10] Vapnik, V. N. The Nature of Statistical Learning Theory[M]. Springer, 1995.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家“十五”科技攻关计划重点项目(A3480266);广东省自然科学基金项目(B6480598)
{{custom_fund}}
PDF(375 KB)
