汉语依存句法分析关键技术研究
|
摘要
依存句法分析的目标是分析输入句子的句法结构，将词语序列转化为树状的依存结构。一条依存弧两个词语构成搭配关系，依存弧上的标签表示搭配的具体类型，如主语、宾语、状语等。在不同语法体系中，依存语法以其形式简洁、易于标注、便于应用等优点，逐渐受到研究人员的重视。计算自然语言学习国际会议（CoNLL）联合举办的公开评测任务从2006年至2009年连续四年关注面向多语言的依存句法分析，大大推动了依存句法分析的发展。依存句法分析也越来越广泛的应用于机器翻译、问答系统、文本挖掘、信息检索等。
依存句法分析的研究工作旨在提高依存分析的准确率和效率。高准确率的分析结果可以为上层应用提供更可信的句法结构。随着互联网数据的迅速膨胀，上层应用系统需要迅速处理海量的信息，因此效率对于依存句法分析的应用也很关键。本文的研究内容涵盖这两个问题，包括以下四个方面。
1.提出了基于柱搜索和标点切分的快速高阶依存句法分析方法。针对前人提出的面向高阶依存句法分析模型的动态规划解码算法时间复杂度高的问题，本文提出使用柱搜索的近似解码算法，一方面允许模型可以方便的融入丰富的高阶句法子树特征，另一方面保证较低的时间复杂度，我们实现的基于柱搜索的高阶依存句法分析系统在CoNLL2009年多语依存句法分析和语义角色标注联合评测任务上取得了优异的成绩。进而，我们针对汉语的特点，提出一种利用标点符号进行长句切分的二阶段依存句法分析方法，进一步提高依存句法分析模型处理长句时的效率。实验证明，这种方法可以大幅度提高依存句法分析的速度，长句子的句法分析准确率也有提高。
2.提出了汉语词性标注和依存句法分析联合模型。由于缺少词语的形态变化信息，汉语词性标注和其他语言如英语相比，准确率较低。这对对汉语依存句法分析带来严重的错误蔓延问题。实验表明使用自动词性时依存句法分析准确率比使用正确词性时低大约6%。对此，本文提出并深入系统的研究了汉语词性标注和依存句法分析联合模型。首先，我们扩展了前人提出的面向依存句法分析的解码算法，提出了相应的面向联合模型的基于动态规划的解码算法。并且，为了解决联合解码算法的时间复杂度过高的问题，我们又提出了一种有效地基于边缘概率的词性裁剪方法。实验结果表明联合模型可以提高词性和句法准确率。深入的错误分析表明联合模型可以帮助消解句法敏感的词性歧义。
3.提出了面向联合模型的分离被动进取训练算法。词性标注和依存句法分析联合模型中句法特征占据主导地位，导致词性特征无法贡献其消歧作用。对此，本文提出一种面向词性标注和依存句法分析联合模型的训练算法。算法分别不同的步长对词性特征和句法特征的权重进行更新。和传统的平均感知器和被动进取训练算法相比，分离被动进取训练算法可以很自然的增大词性特征的权重，从而更好的平衡联合模型中词性特征和句法特征的消歧作用。实验发现，我们的基于分离被动进取训练算法的联合模型在汉语和英语数据上都可以取得最好的词性和句法准确率。
4.提出了基于准同步文法的多树库融合方法。汉语存在多个异构树库，而利用多个树库以提高依存句法分析准确率是一个非常有吸引力的课题。本文提出一种基于准同步文法的多树库融合方法，充分利用标注规范不同的多个单语树库，以提高句法分析准确率。我们设计了丰富的转换模式来刻画不同标注规范间的对应规律，然后基于这些转化模式形成准同步文法特征，从而增强基准依存句法分析模型。准同步文法特征用来指导句法模型做出更好的决策，并且可以很自然的融入到基于图的句法分析解码算法中。实验结果表明，我们的方法可以充分利用源树库的知识。从而提高句法模型在目标树库上的准确率。
总之，本文针对汉语特点，深入研究了基于标点的快速高阶依存句法分析方法、词性标注和依存句法分析的联合模型和多树库融合问题，大大提高了汉语依存句法分析处理实际文本数据的效率和准确率。本研究取得了一些初步的成果。我们期待这些研究成果可以进一步推动自然语言处理领域和其他上层应用如机器翻译、信息抽取的发展。
Dependency parsing aims to analyze the syntactic structure of a given sentence, andconverts the word sequence into a tree. A dependency consists of two words, and the oneword modifies the other one. The label of the dependency represents the specific relationbetween the two words, such as subject, object, adverbial modifier, and so on. Amongdiferent syntax formulizations, the dependency grammar has gained more and more in-terest in parsing community due to its characteristics:(1) representation simplicity,(2)easy to annotate,(3) easy to make use of. The international conference of ComputationalNatural Language Learning (CoNLL) has organized shared tasks on multilingual depen-dency parsing from2006to2009, which largely promotes the research on dependencyparsing. Meanwhile, dependency parsing has been more extensively applied to machinetranslation, question answering, text mining, information retrieval, and so on.
Research on dependency parsing has two important goals. One goal is to improve theparsing accuracy, while the other is to improve the parsing efciency. Accurate parsingresults can provide reliable syntactic structures for higher-level applications. Along withthe quick growth of the web data, higher-level applications need to process a large amountof information in some limited time. Therefore, parsing efciency is also important. Thisthesis covers the two issues and consists of four parts.
1. We propose a fast high-order dependency parsing method based on beam searchand punctuation. The previously proposed decoding algorithm for high-order dependencyparsing is based on dynamic programming and has high time complexity. To address thisissue, we propose a beam-search based decoding algorithm which on one hand allows themodel to incorporate rich high-order syntactic features, and on the other hand is able tofind the approximate optimal parse tree under lower time complexity. Our beam searchbased high-order dependency parser attended the CoNLL2009shared task on multilin-gual dependency parsing and semantic role labeling and achieved good results. To fur-ther improve the parsing efciency for long sentences, we analyze the characteristics ofChinese and propose to use punctuation to segment an input sentence into several subsen-tences and then apply two-stage dependency parsing. Experimental results show that thispunctuation-based two-stage parsing method can largely improve the parsing speed for long sentences. Meanwhile, the parsing accuracy on long sentences is also substantiallyincreased.
2. We propose joint models for Chinese POS tagging and dependency parsing. Dueto little morphological changes, Chinese POS tagging accuracy is much lower than otherlanguages like English. This leads to severe error propagation for Chinese dependencyparsing. Our experiments show that parsing accuracy drops by about6%when replacingmanual POS tags of the input sentence with automatic ones generated by a state-of-the-art statistical POS tagger. To address this issue, this paper proposes a solution by jointlyoptimizing POS tagging and dependency parsing in a unique model.1) We propose forour joint models several dynamic programming based decoding algorithms by extendingthe decoding algorithms for dependency parsing.2) A novel and efective pruning strategybased on marginal probabilities is presented to reduce the search space of candidate POStags. Experimental results show that our joint models significantly improve both the state-of-the-art tagging and parsing accuracies. Detailed analysis shows that the joint methodcan help resolve syntax-sensitive POS ambiguities.
3. We propose a separately passive-aggressive training algorithm for joint models.Joint models for POS tagging and dependency parsing are dominated by syntactic fea-tures. As a result, the POS features fails to fully contribute their disambiguation power.To solve this issue, we propose a separately passive-aggressive learning algorithm (SPA),which is designed to separately update the POS features weights and the syntactic featureweights with diferent update steps under the joint optimization framework. Comparedwith the traditional training algorithms averaged perceptron (AP) and passive aggressive(PA), SPA can naturally raise the weights of the POS features, and therefore better balancethe discriminative power of the POS and syntactic features of the joint models. Experi-mental results show that our joint models trained with SPA achieve the best tagging andparsing accuracy on both Chinese and English datasets.
4. We propose a new multiple treebank exploitation method for dependency pars-ing with quasi-synchronous grammar (QG). There exist multiple treebanks of diferentannotation styles for Chinese, and it is attractive to exploit multiple treebanks to improvethe parsing accuracy. We present a simple and efective framework based on QG for ex-ploiting multiple monolingual treebanks with diferent annotation guidelines for parsing.Several types of transformation patterns (TP) are designed to capture the systematic an-notation inconsistencies among diferent treebanks. Based on such TPs, we design QG features to augment the baseline parsing models. The QG features can guide the parsingmodel to make better decisions, and they naturally fit into the decoding algorithms ofthe baseline graph-based parsing models. Experimental results show that our method canefectively exploit the knowledge of the source treebank, and significantly improve theparsing accuracy on the target treebank.
In conclusion, based on the characteristics of Chinese, this thesis conducts thoroughstudy on fast high-order dependency parsing using punctuation, joint POS tagging anddependency parsing, and multiple treebank exploitation, and substantially improve the ef-ficiency and accuracy of dependency parsing on real-world texts. We have accomplishedseveral primitive achievements, which we hope can further motivate the progress of nat-ural language processing and other high-level applications like machine translation andinformation retrieval.
依存句法分析的研究工作旨在提高依存分析的准确率和效率。高准确率的分析结果可以为上层应用提供更可信的句法结构。随着互联网数据的迅速膨胀，上层应用系统需要迅速处理海量的信息，因此效率对于依存句法分析的应用也很关键。本文的研究内容涵盖这两个问题，包括以下四个方面。
1.提出了基于柱搜索和标点切分的快速高阶依存句法分析方法。针对前人提出的面向高阶依存句法分析模型的动态规划解码算法时间复杂度高的问题，本文提出使用柱搜索的近似解码算法，一方面允许模型可以方便的融入丰富的高阶句法子树特征，另一方面保证较低的时间复杂度，我们实现的基于柱搜索的高阶依存句法分析系统在CoNLL2009年多语依存句法分析和语义角色标注联合评测任务上取得了优异的成绩。进而，我们针对汉语的特点，提出一种利用标点符号进行长句切分的二阶段依存句法分析方法，进一步提高依存句法分析模型处理长句时的效率。实验证明，这种方法可以大幅度提高依存句法分析的速度，长句子的句法分析准确率也有提高。
2.提出了汉语词性标注和依存句法分析联合模型。由于缺少词语的形态变化信息，汉语词性标注和其他语言如英语相比，准确率较低。这对对汉语依存句法分析带来严重的错误蔓延问题。实验表明使用自动词性时依存句法分析准确率比使用正确词性时低大约6%。对此，本文提出并深入系统的研究了汉语词性标注和依存句法分析联合模型。首先，我们扩展了前人提出的面向依存句法分析的解码算法，提出了相应的面向联合模型的基于动态规划的解码算法。并且，为了解决联合解码算法的时间复杂度过高的问题，我们又提出了一种有效地基于边缘概率的词性裁剪方法。实验结果表明联合模型可以提高词性和句法准确率。深入的错误分析表明联合模型可以帮助消解句法敏感的词性歧义。
3.提出了面向联合模型的分离被动进取训练算法。词性标注和依存句法分析联合模型中句法特征占据主导地位，导致词性特征无法贡献其消歧作用。对此，本文提出一种面向词性标注和依存句法分析联合模型的训练算法。算法分别不同的步长对词性特征和句法特征的权重进行更新。和传统的平均感知器和被动进取训练算法相比，分离被动进取训练算法可以很自然的增大词性特征的权重，从而更好的平衡联合模型中词性特征和句法特征的消歧作用。实验发现，我们的基于分离被动进取训练算法的联合模型在汉语和英语数据上都可以取得最好的词性和句法准确率。
4.提出了基于准同步文法的多树库融合方法。汉语存在多个异构树库，而利用多个树库以提高依存句法分析准确率是一个非常有吸引力的课题。本文提出一种基于准同步文法的多树库融合方法，充分利用标注规范不同的多个单语树库，以提高句法分析准确率。我们设计了丰富的转换模式来刻画不同标注规范间的对应规律，然后基于这些转化模式形成准同步文法特征，从而增强基准依存句法分析模型。准同步文法特征用来指导句法模型做出更好的决策，并且可以很自然的融入到基于图的句法分析解码算法中。实验结果表明，我们的方法可以充分利用源树库的知识。从而提高句法模型在目标树库上的准确率。
总之，本文针对汉语特点，深入研究了基于标点的快速高阶依存句法分析方法、词性标注和依存句法分析的联合模型和多树库融合问题，大大提高了汉语依存句法分析处理实际文本数据的效率和准确率。本研究取得了一些初步的成果。我们期待这些研究成果可以进一步推动自然语言处理领域和其他上层应用如机器翻译、信息抽取的发展。
Dependency parsing aims to analyze the syntactic structure of a given sentence, andconverts the word sequence into a tree. A dependency consists of two words, and the oneword modifies the other one. The label of the dependency represents the specific relationbetween the two words, such as subject, object, adverbial modifier, and so on. Amongdiferent syntax formulizations, the dependency grammar has gained more and more in-terest in parsing community due to its characteristics:(1) representation simplicity,(2)easy to annotate,(3) easy to make use of. The international conference of ComputationalNatural Language Learning (CoNLL) has organized shared tasks on multilingual depen-dency parsing from2006to2009, which largely promotes the research on dependencyparsing. Meanwhile, dependency parsing has been more extensively applied to machinetranslation, question answering, text mining, information retrieval, and so on.
Research on dependency parsing has two important goals. One goal is to improve theparsing accuracy, while the other is to improve the parsing efciency. Accurate parsingresults can provide reliable syntactic structures for higher-level applications. Along withthe quick growth of the web data, higher-level applications need to process a large amountof information in some limited time. Therefore, parsing efciency is also important. Thisthesis covers the two issues and consists of four parts.
1. We propose a fast high-order dependency parsing method based on beam searchand punctuation. The previously proposed decoding algorithm for high-order dependencyparsing is based on dynamic programming and has high time complexity. To address thisissue, we propose a beam-search based decoding algorithm which on one hand allows themodel to incorporate rich high-order syntactic features, and on the other hand is able tofind the approximate optimal parse tree under lower time complexity. Our beam searchbased high-order dependency parser attended the CoNLL2009shared task on multilin-gual dependency parsing and semantic role labeling and achieved good results. To fur-ther improve the parsing efciency for long sentences, we analyze the characteristics ofChinese and propose to use punctuation to segment an input sentence into several subsen-tences and then apply two-stage dependency parsing. Experimental results show that thispunctuation-based two-stage parsing method can largely improve the parsing speed for long sentences. Meanwhile, the parsing accuracy on long sentences is also substantiallyincreased.
2. We propose joint models for Chinese POS tagging and dependency parsing. Dueto little morphological changes, Chinese POS tagging accuracy is much lower than otherlanguages like English. This leads to severe error propagation for Chinese dependencyparsing. Our experiments show that parsing accuracy drops by about6%when replacingmanual POS tags of the input sentence with automatic ones generated by a state-of-the-art statistical POS tagger. To address this issue, this paper proposes a solution by jointlyoptimizing POS tagging and dependency parsing in a unique model.1) We propose forour joint models several dynamic programming based decoding algorithms by extendingthe decoding algorithms for dependency parsing.2) A novel and efective pruning strategybased on marginal probabilities is presented to reduce the search space of candidate POStags. Experimental results show that our joint models significantly improve both the state-of-the-art tagging and parsing accuracies. Detailed analysis shows that the joint methodcan help resolve syntax-sensitive POS ambiguities.
3. We propose a separately passive-aggressive training algorithm for joint models.Joint models for POS tagging and dependency parsing are dominated by syntactic fea-tures. As a result, the POS features fails to fully contribute their disambiguation power.To solve this issue, we propose a separately passive-aggressive learning algorithm (SPA),which is designed to separately update the POS features weights and the syntactic featureweights with diferent update steps under the joint optimization framework. Comparedwith the traditional training algorithms averaged perceptron (AP) and passive aggressive(PA), SPA can naturally raise the weights of the POS features, and therefore better balancethe discriminative power of the POS and syntactic features of the joint models. Experi-mental results show that our joint models trained with SPA achieve the best tagging andparsing accuracy on both Chinese and English datasets.
4. We propose a new multiple treebank exploitation method for dependency pars-ing with quasi-synchronous grammar (QG). There exist multiple treebanks of diferentannotation styles for Chinese, and it is attractive to exploit multiple treebanks to improvethe parsing accuracy. We present a simple and efective framework based on QG for ex-ploiting multiple monolingual treebanks with diferent annotation guidelines for parsing.Several types of transformation patterns (TP) are designed to capture the systematic an-notation inconsistencies among diferent treebanks. Based on such TPs, we design QG features to augment the baseline parsing models. The QG features can guide the parsingmodel to make better decisions, and they naturally fit into the decoding algorithms ofthe baseline graph-based parsing models. Experimental results show that our method canefectively exploit the knowledge of the source treebank, and significantly improve theparsing accuracy on the target treebank.
In conclusion, based on the characteristics of Chinese, this thesis conducts thoroughstudy on fast high-order dependency parsing using punctuation, joint POS tagging anddependency parsing, and multiple treebank exploitation, and substantially improve the ef-ficiency and accuracy of dependency parsing on real-world texts. We have accomplishedseveral primitive achievements, which we hope can further motivate the progress of nat-ural language processing and other high-level applications like machine translation andinformation retrieval.
引文
[1] Collins M. Head-driven statistical models for natural language parsing[D]. Uni-viersity of Pennsylvania,1999.
[2] Mel’cˇuk I A. Dependency Syntax: Theory and Practice[M]. State University ofNew York Press,1988.
[3] Pollard C, Sag I A. Head-Driven Phrase Structure Grammar[M]. Chicago: Univer-sity of Chicago Press and Stanford: CSLI Publications,1994.
[4]马金山.基于统计方法的汉语依存句法分析研究[D].哈尔滨工业大学,2007.
[5] Che W, Spitkovsky V, Liu T. A Comparison of Chinese Parsers for StanfordDependencies[C]. Proceedings of the50th Annual Meeting of the Association forComputational Linguistics.2012:11–16.
[6]车万翔.基于核方法的语义角色标注研究[D].哈尔滨工业大学,2008.
[7]李军辉.中文句法语义分析及其联合学习机制研究[D].苏州大学,2010.
[8] HajicˇJ, Ciaramita M, Johansson R, et al. The CoNLL-2009Shared Task: Syntacticand Semantic Dependencies in Multiple Languages[C]. Proceedings of CoNLL2009.2009.
[9] Buchholz S, Marsi E. CoNLL-X Shared Task on Multilingual Dependency Pars-ing[C]. In Proc. of CoNLL.2006:149–164.
[10] Nivre J, Hall J, Ku¨bler S, et al. The CoNLL2007Shared Task on DependencyParsing[C]. Proceedings of the CoNLL Shared Task Session of EMNLP-CoNLL2007.2007:915–932.
[11] Surdeanu M, Johansson R, Meyers A, et al. The CoNLL-2008Shared Task onJoint Parsing of Syntactic and Semantic Dependencies[C]. CoNLL-2008.2008.
[12] Li Z, Zhang M, Che W, et al. Joint Models for Chinese POS Tagging and Depen-dency Parsing[C]. EMNLP2011.2011:1180–1191.
[13] Hatori J, Matsuzaki T, Miyao Y, et al. Incremental Joint POS Tagging and Depen-dency Parsing in Chinese[C]. Proceedings of5th International Joint Conference onNatural Language Processing.2011:1216–1224.
[14] Bohnet B, Nivre J. A Transition-Based System for Joint Part-of-Speech Tag-ging and Labeled Non-Projective Dependency Parsing[C]. Proceedings of the2012Joint Conference on Empirical Methods in Natural Language Processing and Com-putational Natural Language Learning.2012:1455–1465.
[15] Hatori J, Matsuzaki T, Miyao Y, et al. Incremental Joint Approach to WordSegmentation, POS Tagging, and Dependency Parsing in Chinese[C]. Proceed-ings of the50th Annual Meeting of the Association for Computational Linguistics.2012:1045–1053.
[16] Li Z, Zhou G. Unified Dependency Parsing of Chinese Morphological and Syntac-tic Structures[C]. Proceedings of the2012Joint Conference on Empirical Methodsin Natural Language Processing and Computational Natural Language Learning.2012:1445–1454.
[17] Qian X, Liu Y. Joint Chinese Word Segmentation, POS Tagging and Parsing[C].Proceedings of the2012Joint Conference on Empirical Methods in Natural Lan-guage Processing and Computational Natural Language Learning.2012:501–511.
[18] Kubler S, McDonald R, Nivre J. Dependency Parsing (Synthesis Lectures OnHuman Language Technologies)[M]. Morgan and Claypool Publishers,2009.
[19] McDonald R, Crammer K, Pereira F. Online Large-Margin Training of Dependen-cy Parsers[C]. Proceedings of ACL2005.2005:91–98.
[20] McDonald R, Pereira F. Online Learning of Approximate Dependency ParsingAlgorithms[C]. Proceedings of EACL2006.2006.
[21] Carreras X. Experiments with a Higher-Order Projective Dependency Parser[C].Proceedings of EMNLP/CoNLL.2007:141–150.
[22] Koo T, Collins M. Efcient Third-Order Dependency Parsers[C]. Proceedingsof the48th Annual Meeting of the Association for Computational Linguistics.2010:1–11.
[23] Eisner J. Bilexical Grammars and a Cubic-Time Probabilistic Parser[C]. Proceed-ings of the5th International Workshop on Parsing Technologies (IWPT).1997:54–65.
[24] Eisner J. Bilexical Grammars and Their Cubic-Time Parsing Algorithms[M]..Bunt H C, Nijholt A. Advances in Probabilistic and Other Parsing Technologies.Kluwer Academic Publishers,2000:29–62.
[25] McDonald R. Discriminative Training and Spanning Tree Algorithms for Depen-dency Parsing[D]. University of Pennsylvania,2006.
[26] Collins M. Discriminative Training Methods for Hidden Markov Models: Theoryand Experiments with Perceptron Algorithms[C]. Proceedings of EMNLP2002.2002.
[27] Crammer K, Dekel O, Keshet J, et al. Online passive aggressive algorithms[C].Procedings of NIPS2003.2003.
[28] Crammer K, Singer Y. Ultraconservative Online Algorithms for Multiclass Prob-lems[J]. Journal of Machine Learning Research,2001,3.
[29] Smith D A, Eisner J. Dependency Parsing by Belief Propagation[C]. Proceedingsof EMNLP2008.2008:145–156.
[30] Riedel S, Clarke J. Incremental integer linear programming for non-projectivedependency parsing[C]. In EMNLP.2006:129–137.
[31] Martins A, Smith N, Xing E. Concise Integer Linear Programming Formulationsfor Dependency Parsing[C]. Proceedings of ACL/IJCNLP2009.2009:342–350.
[32] Yamada H, Matsumoto Y. Statistical dependency analysis with support vectormachines[C]. Proceedings of IWPT2003.2003:195–206.
[33] Nivre J. An Efcient Algorithm for Projective Dependency Parsing[C]. Pro-ceedings of the8th International Workshop on Parsing Technologies (IWPT).2003:149–160.
[34] Huang L, Sagae K. Dynamic Programming for Linear-Time Incremental Pars-ing[C]. Proceedings of the48th Annual Meeting of the Association for Computa-tional Linguistics.2010:1077–1086.
[35] Zhang Y, Nivre J. Transition-based Dependency Parsing with Rich Non-localFeatures[C]. Proceedings of the49th Annual Meeting of the Association for Com-putational Linguistics: Human Language Technologies.2011:188–193.
[36] Zhang Y, Clark S. A Tale of Two Parsers: Investigating and Combining Graph-based and Transition-based Dependency Parsing[C]. Proceedings of the2008Con-ference on Empirical Methods in Natural Language Processing.2008:562–571.
[37] Nivre J. Algorithms for deterministic incremental dependency parsing[C]. Com-putational Linguistics.2008，34:513–553.
[38] Huang L, Jiang W, Liu Q. Bilingually-Constrained (Monolingual) Shift-ReduceParsing[C]. Proceedings of the2009Conference on Empirical Methods in NaturalLanguage Processing.2009:1222–1231.
[39] Nivre J, Hall J. Maltparser: A language-independent system for data-driven de-pendency parsing[C]. In Proc. of the Fourth Workshop on Treebanks and LinguisticTheories.2005:13–95.
[40] Duan X, Xu J Z B. Probabilistic Parsing Action Models for Multi-Lingual De-pendency Parsing[C]. Proceedings of the CoNLL Shared Task Session of EMNLP-CoNLL2007.2007:940–946.
[41] Johansson R, Nugues P. Incremental Dependency Parsing Using Online Learn-ing[C]. Proceedings of the CoNLL Shared Task Session of EMNLP-CoNLL2007.2007:1134–1138.
[42] McDonald R, Nivre J. Characterizing the Errors of Data-Driven DependencyParsing Models[C]. Proceedings of EMNLP-CoNLL2007.2007:122–131.
[43] Sagae K, Lavie A. Parser Combination by Reparsing[C]. Proceedings of theHuman Language Technology Conference of the NAACL, Companion Volume.2006:129–132.
[44] Nivre J, McDonald R. Integrating Graph-Based and Transition-Based DependencyParsers[C]. Proceedings of ACL2008.2008:950–958.
[45] Koo T, Carreras X, Collins M. Simple Semi-supervised Dependency Parsing[C].Proceedings of ACL-08: HLT.2008:595–603.
[46] Chen W, Kazama J, Uchimoto K, et al. Improving Dependency Parsing with Sub-trees from Auto-Parsed Data[C]. Proceedings of the2009Conference on EmpiricalMethods in Natural Language Processing.2009:570–579.
[47] McClosky D, Charniak E, Johnson M. Efective Self-Training for Parsing[C].Proceedings of the Human Language Technology Conference of the NAACL.2006:152–159.
[48] McClosky D, Charniak E, Johnson M. Reranking and Self-Training for ParserAdaptation[C]. Proceedings of the21st International Conference on Computation-al Linguistics and44th Annual Meeting of the Association for Computational Lin-guistics.2006:337–344.
[49] McClosky D, Charniak E, Johnson M. When is Self-Training Efective for Pars-ing?[C]. Proceedings of the22nd International Conference on Computational Lin-guistics (Coling2008).2008:561–568.
[50] Charniak E, Johnson M. Coarse-to-Fine n-Best Parsing and MaxEnt DiscriminativeReranking[C]. Proceedings of ACL-05.2005:173–180.
[51] Chu Y J, Liu T H. On the shortest arborescence of a directed graph[J]. ScienceSinica,1965,14:1396–1400.
[52] McDonald R, Pereira F, Ribarov K, et al. Non-Projective Dependency Parsing us-ing Spanning Tree Algorithms[C]. Proceedings of Human Language TechnologyConference and Conference on Empirical Methods in Natural Language Process-ing.2005:523–530.
[53] Nivre J, Nilsson J. Pseudo-Projective Dependency Parsing[C]. Proceedings of the43rd Annual Meeting of the Association for Computational Linguistics (ACL05).2005:99–106.
[54] Nivre J. Non-Projective Dependency Parsing in Expected Linear Time[C]. Pro-ceedings of the Joint Conference of the47th Annual Meeting of the ACL and the4th International Joint Conference on Natural Language Processing of the AFNLP.2009:351–359.
[55] McClosky D, Charniak E, Johnson M. Automatic Domain Adaptation for Pars-ing[C]. Human Language Technologies: The2010Annual Conference of the NorthAmerican Chapter of the Association for Computational Linguistics.2010:28–36.
[56]段湘煜,赵军,徐波.基于动作建模的中文依存句法分析[J].中文信息学报,2007,21(5).
[57]辛霄,范士喜,王轩,王晓龙.基于最大熵的依存句法分析[J].中文信息学报,2009,23(2).
[58]鉴萍,宗成庆.基于序列标注模型的分层式依存句法分析方法[J].中文信息学报,2010,24(6).
[59]计峰,邱锡鹏.基于序列标注的中文依存句法分析方法[J].计算机应用与软件,2009,26(10).
[60] Duan X, Zhao J,, et al. Probabilistic models for action-based Chinese dependencyparsing[C]. Proceedings of ECML/ECPPKDD.2007.
[61] Titov I, Henderson J. Fast and Robust Multilingual Dependency Parsing witha Generative Latent Variable Model[C]. Proceedings of the CoNLL Shared TaskSession of EMNLP-CoNLL2007.2007:947–951.
[62] Eisner J. Efcient Normal-Form Parsing for Combinatory Categorial Grammar[C].Proceedings of the34th Annual Meeting of the Association for Computational Lin-guistics (ACL).1996:79–86.
[63] Chen W, Kazama J, Tsuruoka Y, et al. Improving Graph-based Dependency Parsingwith Decision History[C]. Coling2010: Posters.2010:126–134.
[64] Zhang H, McDonald R. Generalized Higher-Order Dependency Parsing withCube Pruning[C]. Proceedings of the2012Joint Conference on Empirical Method-s in Natural Language Processing and Computational Natural Language Learning.2012:320–331.
[65] Nunberg G. The linguistics of punctuation[M]. Lecture Notes, no.18. Stanford,CA: Center for the Study of Language and Information,1990.
[66] Collins M. Head-Driven Statistical Models for Natural Language Parsing[C].Computational Linguistics.2003，29:589–637.
[67] White M, Rajkumar R. A More Precise Analysis of Punctuation for Broad-Coverage Surface Realization with CCG[C]. Coling2008: workshop on GrammarEngineering Across Frameworks.2008:17–24.
[68] xun Jin M, Kim M Y, Kim D, et al. Segmentation of Chinese Long SentencesUsing Commas[C]. ACL SIGHAN Workshop2004.2004.
[69] Li X, Zong C, Hu R. A Hierarchical Parsing Approach with Punctuation Processingfor Long Chinese Sentences[C]. IJCNLP2005: Companion Volume.2005.
[70]毛奇,连乐新,周文翠,袁春风.基于标点符号分割的汉语句法分析算法[J].中文信息学报,2007,21(2).
[71] Bohnet B. Efcient Parsing of Syntactic and Semantic Dependency Structures[C].Proceedings of the Thirteenth Conference on Computational Natural LanguageLearning (CoNLL2009): Shared Task.2009:67–72.
[72] Bohnet B. Top Accuracy and Fast Dependency Parsing is not a Contradiction[C].Proceedings of the23rd International Conference on Computational Linguistics(Coling2010).2010:89–97.
[73] Huang L, Chiang D. Better k-best Parsing[C]. Proceedings of IWPT2005.2005.
[74] Gesmundo A, Henderson J, Merlo P, et al. A Latent Variable Model of Syn-chronous Syntactic-Semantic Parsing for Multiple Languages[C]. Proceedings ofCoNLL2009: Shared Task.2009:37–42.
[75] Che W, Li Z, Li Y, et al. Multilingual Dependency-based Syntactic and SemanticParsing[C]. Proceedings of CoNLL2009: Shared Task.2009:49–54.
[76] Xue N, Xia F, Chiou F D, et al. The Penn Chinese TreeBank: Phrase StructureAnnotation of a Large Corpus[C]. Natural Language Engineering.2005，11:207–238.
[77] Zhang Y, Clark S. Joint Word Segmentation and POS Tagging Using a SinglePerceptron[C]. Proceedings of ACL-08: HLT.2008:888–896.
[78] Jiang W, Huang L, Liu Q, et al. A Cascaded Linear Model for Joint ChineseWord Segmentation and Part-of-Speech Tagging[C]. Proceedings of ACL-08: HLT.2008:897–904.
[79] Kruengkrai C, Uchimoto K, Kazama J, et al. An Error-Driven Word-CharacterHybrid Model for Joint Chinese Word Segmentation and POS Tagging[C]. Pro-ceedings of the Joint Conference of the47th Annual Meeting of the ACL and the4th International Joint Conference on Natural Language Processing of the AFNLP.2009:513–521.
[80] Toutanova K, Cherry C. A global model for joint lemmatization and part-of-speechprediction[C]. Proceedings of the Joint Conference of the47th Annual Meeting ofthe ACL and the4th International Joint Conference on Natural Language Process-ing of the AFNLP.2009:486–494.
[81] Cohen S B, Smith N A. Joint Morphological and Syntactic Disambiguation[C].Proceedings of EMNLP-CoNLL2007.2007:208–217.
[82] Goldberg Y, Tsarfaty R. A Single Generative Model for Joint MorphologicalSegmentation and Syntactic Parsing[C]. Proceedings of ACL-08: HLT.2008:371–379.
[83] Finkel J R, Manning C D. Joint Parsing and Named Entity Recognition[C].Proceedings of Human Language Technologies: The2009Annual Conferenceof the North American Chapter of the Association for Computational Linguistics.2009:326–334.
[84] Li J, Zhou G, Ng H T. Joint Syntactic and Semantic Parsing of Chinese[C]. Pro-ceedings of the48th Annual Meeting of the Association for Computational Lin-guistics.2010:1108–1117.
[85] Che W, Liu T. Jointly Modeling WSD and SRL with Markov Logic[C]. Proceed-ings of the23rd International Conference on Computational Linguistics (Coling2010).2010:161–169.
[86] Dyer C. Using a maximum entropy model to build segmentation lattices forMT[C]. Proceedings of Human Language Technologies: The2009Annual Con-ference of the North American Chapter of the Association for Computational Lin-guistics.2009:406–414.
[87] Xiao X, Liu Y, Hwang Y, et al. Joint Tokenization and Translation[C]. Proceedingsof the23rd International Conference on Computational Linguistics (Coling2010).2010:1200–1208.
[88] Liu Y, Liu Q. Joint Parsing and Translation[C]. Proceedings of the23rd Interna-tional Conference on Computational Linguistics (Coling2010).2010:707–715.
[89] McDonald R, Lerman K, Pereira F. Multilingual Dependency Analysis with a Two-Stage Discriminative Parser[C]. Proceedings of the Tenth Conference on Compu-tational Natural Language Learning (CoNLL-X).2006:216–220.
[90] Lee J, Naradowsky J, Smith D A. A Discriminative Model for Joint Morpholog-ical Disambiguation and Dependency Parsing[C]. Proceedings of the49th Annu-al Meeting of the Association for Computational Linguistics: Human LanguageTechnologies.2011:885–894.
[91] Rush A M, Sontag D, Collins M, et al. On Dual Decomposition and Linear Pro-gramming Relaxations for Natural Language Processing[C]. Proceedings of the2010Conference on Empirical Methods in Natural Language Processing.2010:1–11.
[92] Auli M, Lopez A. A Comparison of Loopy Belief Propagation and Dual De-composition for Integrated CCG Supertagging and Parsing[C]. Proceedings of the49th Annual Meeting of the Association for Computational Linguistics: HumanLanguage Technologies.2011:470–480.
[93] Ratnaparkhi A. A Maximum Entropy Model for Part-Of-Speech Tagging[C]. Pro-ceedings of EMNLP1996.1996.
[94] Laferty J, McCallum A, Pereira F. Conditional Random Fields: ProbabilisticModels for Segmenting and Labeling Sequence Data[C]. Proceedings of ICML2001.2001:282–289.
[95] Collins M, Globerson A, Koo T, et al. Exponentiated Gradient Algorithms forConditional Random Fields and Max-Margin Markov Networks[J]. JMLR,2008,9:1775–1822.
[96] Eisner J. Three New Probabilistic Models for Dependency Parsing: An Explo-ration[C]. Proceedings of COLING1996.1996:340–345.
[97] Zhang Y, Clark S. Syntactic processing using the generalized perceptron and beamsearch[J]. Computational Linguistics,2011,37(1):105–151.
[98] Liu X, Zhou M, Zhou X, et al. Joint Inference of Named Entity Recognitionand Normalization for Tweets[C]. Proceedings of the50th Annual Meeting of theAssociation for Computational Linguistics.2012:526–535.
[99] Constant M, Sigogne A, Watrin P. Discriminative Strategies to Integrate Mul-tiword Expression Recognition and Parsing[C]. Proceedings of the50th AnnualMeeting of the Association for Computational Linguistics (Volume1: Long Paper-s).2012:204–212.
[100] Minkov E, Zettlemoyer L. Discriminative Learning for Joint Template Filling[C].Proceedings of the50th Annual Meeting of the Association for Computational Lin-guistics.2012:845–853.
[101] Petrov S, Klein D. Improved Inference for Unlexicalized Parsing[C]. Proceedingsof NAACL2007.2007.
[102] Johansson R, Nugues P. Dependency-based Semantic Role Labeling of Prop-Bank[C]. EMNLP-2008.2008.
[103] Bansal M, Klein D. Web-Scale Features for Full-Scale Parsing[C]. Proceedings ofthe49th Annual Meeting of the Association for Computational Linguistics: HumanLanguage Technologies.2011:693–702.
[104] Zhou G, Zhao J, Liu K, et al. Exploiting Web-Derived Selectional Preference to Im-prove Statistical Dependency Parsing[C]. Proceedings of the49th Annual Meetingof the Association for Computational Linguistics: Human Language Technologies.2011:1556–1565.
[105] Burkett D, Klein D. Two Languages are Better than One (for Syntactic Parsing)[C].Proceedings of the2008Conference on Empirical Methods in Natural LanguageProcessing.2008:877–886.
[106] Chen W, Kazama J, Torisawa K. Bitext Dependency Parsing with Bilingual SubtreeConstraints[C]. Proceedings of the48th Annual Meeting of the Association forComputational Linguistics.2010:21–29.
[107]王跃龙,姬东鸿.汉语树库综述[J].当代语言学,2009,11(1).
[108] Liu T, Ma J, Li S. Building a Dependency Treebank for Improving Chinese Pars-er[C]. Journal of Chinese Language and Computing.2006，16:207–224.
[109] Chen K J, Luo C C, Chang M C, et al. Sinica treebank: Design criteria, rep-resentational issues and implementation[M]. Kluwer Academic Publishers,2003:231–248.
[110]周强.汉语句法树库标注体系[J].中文信息学报,2004,18(4).
[111] Niu Z Y, Wang H, Wu H. Exploiting Heterogeneous Treebanks for Parsing[C]. Pro-ceedings of the Joint Conference of the47th Annual Meeting of the ACL and the4th International Joint Conference on Natural Language Processing of the AFNLP.2009:46–54.
[112] Jiang W, Huang L, Liu Q. Automatic Adaptation of Annotation Standards: ChineseWord Segmentation and POS Tagging–A Case Study[C]. Proceedings of the JointConference of the47th Annual Meeting of the ACL and the4th International JointConference on Natural Language Processing of the AFNLP.2009:522–530.
[113] Smith D A, Eisner J. Parser Adaptation and Projection with Quasi-SynchronousGrammar Features[C]. Proceedings of the2009Conference on Empirical Methodsin Natural Language Processing.2009:822–831.
[114] Wang J N, Chang J S, Su K Y. An Automatic Treebank Conversion Algorithm forCorpus Sharing[C]. Proceedings of the32nd Annual Meeting of the Associationfor Computational Linguistics.1994:248–254.
[115] Collins M, Ramshaw L, Hajic J, et al. A Statistical Parser for Czech[C]. ACL1999.1999:505–512.
[116] Xia F, Palmer M. Converting Dependency Structures to Phrase Structures[C]. InProceedings of HLT2001.2001:1–5.
[117] Xia F, Bhatt R, Rambow O, et al. Towards a Multi-Representational Treebank[C].In Proceedings of the7th International Workshop on Treebanks and Linguistic The-ories.2008.
[118] Li Z, Che W, Liu T. A Study on Constituent-to-Dependency Conversion[J]. Journalof Chinese Information Processing (in Chinese),2008,6(22):14–19.
[119] Martins A F T, Das D, Smith N A, et al. Stacking Dependency Parsers[C].EMNLP’08.2008:157–166.
[120] Smith D, Eisner J. Quasi-Synchronous Grammars: Alignment by Soft Projection ofSyntactic Dependencies[C]. Proceedings on the Workshop on Statistical MachineTranslation.2006:23–30.
[121] Gimpel K, Smith N A. Quasi-Synchronous Phrase Dependency Grammars for Ma-chine Translation[C]. Proceedings of the2011Conference on Empirical Methodsin Natural Language Processing.2011:474–485.
[122] Woodsend K, Lapata M. Learning to Simplify Sentences with Quasi-SynchronousGrammar and Integer Programming[C]. Proceedings of the2011Conference onEmpirical Methods in Natural Language Processing.2011:409–420.
[123] Charniak E. A Maximum-Entropy-Inspired Parser[C]. ANLP’00.2000:132–139.
[2] Mel’cˇuk I A. Dependency Syntax: Theory and Practice[M]. State University ofNew York Press,1988.
[3] Pollard C, Sag I A. Head-Driven Phrase Structure Grammar[M]. Chicago: Univer-sity of Chicago Press and Stanford: CSLI Publications,1994.
[4]马金山.基于统计方法的汉语依存句法分析研究[D].哈尔滨工业大学,2007.
[5] Che W, Spitkovsky V, Liu T. A Comparison of Chinese Parsers for StanfordDependencies[C]. Proceedings of the50th Annual Meeting of the Association forComputational Linguistics.2012:11–16.
[6]车万翔.基于核方法的语义角色标注研究[D].哈尔滨工业大学,2008.
[7]李军辉.中文句法语义分析及其联合学习机制研究[D].苏州大学,2010.
[8] HajicˇJ, Ciaramita M, Johansson R, et al. The CoNLL-2009Shared Task: Syntacticand Semantic Dependencies in Multiple Languages[C]. Proceedings of CoNLL2009.2009.
[9] Buchholz S, Marsi E. CoNLL-X Shared Task on Multilingual Dependency Pars-ing[C]. In Proc. of CoNLL.2006:149–164.
[10] Nivre J, Hall J, Ku¨bler S, et al. The CoNLL2007Shared Task on DependencyParsing[C]. Proceedings of the CoNLL Shared Task Session of EMNLP-CoNLL2007.2007:915–932.
[11] Surdeanu M, Johansson R, Meyers A, et al. The CoNLL-2008Shared Task onJoint Parsing of Syntactic and Semantic Dependencies[C]. CoNLL-2008.2008.
[12] Li Z, Zhang M, Che W, et al. Joint Models for Chinese POS Tagging and Depen-dency Parsing[C]. EMNLP2011.2011:1180–1191.
[13] Hatori J, Matsuzaki T, Miyao Y, et al. Incremental Joint POS Tagging and Depen-dency Parsing in Chinese[C]. Proceedings of5th International Joint Conference onNatural Language Processing.2011:1216–1224.
[14] Bohnet B, Nivre J. A Transition-Based System for Joint Part-of-Speech Tag-ging and Labeled Non-Projective Dependency Parsing[C]. Proceedings of the2012Joint Conference on Empirical Methods in Natural Language Processing and Com-putational Natural Language Learning.2012:1455–1465.
[15] Hatori J, Matsuzaki T, Miyao Y, et al. Incremental Joint Approach to WordSegmentation, POS Tagging, and Dependency Parsing in Chinese[C]. Proceed-ings of the50th Annual Meeting of the Association for Computational Linguistics.2012:1045–1053.
[16] Li Z, Zhou G. Unified Dependency Parsing of Chinese Morphological and Syntac-tic Structures[C]. Proceedings of the2012Joint Conference on Empirical Methodsin Natural Language Processing and Computational Natural Language Learning.2012:1445–1454.
[17] Qian X, Liu Y. Joint Chinese Word Segmentation, POS Tagging and Parsing[C].Proceedings of the2012Joint Conference on Empirical Methods in Natural Lan-guage Processing and Computational Natural Language Learning.2012:501–511.
[18] Kubler S, McDonald R, Nivre J. Dependency Parsing (Synthesis Lectures OnHuman Language Technologies)[M]. Morgan and Claypool Publishers,2009.
[19] McDonald R, Crammer K, Pereira F. Online Large-Margin Training of Dependen-cy Parsers[C]. Proceedings of ACL2005.2005:91–98.
[20] McDonald R, Pereira F. Online Learning of Approximate Dependency ParsingAlgorithms[C]. Proceedings of EACL2006.2006.
[21] Carreras X. Experiments with a Higher-Order Projective Dependency Parser[C].Proceedings of EMNLP/CoNLL.2007:141–150.
[22] Koo T, Collins M. Efcient Third-Order Dependency Parsers[C]. Proceedingsof the48th Annual Meeting of the Association for Computational Linguistics.2010:1–11.
[23] Eisner J. Bilexical Grammars and a Cubic-Time Probabilistic Parser[C]. Proceed-ings of the5th International Workshop on Parsing Technologies (IWPT).1997:54–65.
[24] Eisner J. Bilexical Grammars and Their Cubic-Time Parsing Algorithms[M]..Bunt H C, Nijholt A. Advances in Probabilistic and Other Parsing Technologies.Kluwer Academic Publishers,2000:29–62.
[25] McDonald R. Discriminative Training and Spanning Tree Algorithms for Depen-dency Parsing[D]. University of Pennsylvania,2006.
[26] Collins M. Discriminative Training Methods for Hidden Markov Models: Theoryand Experiments with Perceptron Algorithms[C]. Proceedings of EMNLP2002.2002.
[27] Crammer K, Dekel O, Keshet J, et al. Online passive aggressive algorithms[C].Procedings of NIPS2003.2003.
[28] Crammer K, Singer Y. Ultraconservative Online Algorithms for Multiclass Prob-lems[J]. Journal of Machine Learning Research,2001,3.
[29] Smith D A, Eisner J. Dependency Parsing by Belief Propagation[C]. Proceedingsof EMNLP2008.2008:145–156.
[30] Riedel S, Clarke J. Incremental integer linear programming for non-projectivedependency parsing[C]. In EMNLP.2006:129–137.
[31] Martins A, Smith N, Xing E. Concise Integer Linear Programming Formulationsfor Dependency Parsing[C]. Proceedings of ACL/IJCNLP2009.2009:342–350.
[32] Yamada H, Matsumoto Y. Statistical dependency analysis with support vectormachines[C]. Proceedings of IWPT2003.2003:195–206.
[33] Nivre J. An Efcient Algorithm for Projective Dependency Parsing[C]. Pro-ceedings of the8th International Workshop on Parsing Technologies (IWPT).2003:149–160.
[34] Huang L, Sagae K. Dynamic Programming for Linear-Time Incremental Pars-ing[C]. Proceedings of the48th Annual Meeting of the Association for Computa-tional Linguistics.2010:1077–1086.
[35] Zhang Y, Nivre J. Transition-based Dependency Parsing with Rich Non-localFeatures[C]. Proceedings of the49th Annual Meeting of the Association for Com-putational Linguistics: Human Language Technologies.2011:188–193.
[36] Zhang Y, Clark S. A Tale of Two Parsers: Investigating and Combining Graph-based and Transition-based Dependency Parsing[C]. Proceedings of the2008Con-ference on Empirical Methods in Natural Language Processing.2008:562–571.
[37] Nivre J. Algorithms for deterministic incremental dependency parsing[C]. Com-putational Linguistics.2008，34:513–553.
[38] Huang L, Jiang W, Liu Q. Bilingually-Constrained (Monolingual) Shift-ReduceParsing[C]. Proceedings of the2009Conference on Empirical Methods in NaturalLanguage Processing.2009:1222–1231.
[39] Nivre J, Hall J. Maltparser: A language-independent system for data-driven de-pendency parsing[C]. In Proc. of the Fourth Workshop on Treebanks and LinguisticTheories.2005:13–95.
[40] Duan X, Xu J Z B. Probabilistic Parsing Action Models for Multi-Lingual De-pendency Parsing[C]. Proceedings of the CoNLL Shared Task Session of EMNLP-CoNLL2007.2007:940–946.
[41] Johansson R, Nugues P. Incremental Dependency Parsing Using Online Learn-ing[C]. Proceedings of the CoNLL Shared Task Session of EMNLP-CoNLL2007.2007:1134–1138.
[42] McDonald R, Nivre J. Characterizing the Errors of Data-Driven DependencyParsing Models[C]. Proceedings of EMNLP-CoNLL2007.2007:122–131.
[43] Sagae K, Lavie A. Parser Combination by Reparsing[C]. Proceedings of theHuman Language Technology Conference of the NAACL, Companion Volume.2006:129–132.
[44] Nivre J, McDonald R. Integrating Graph-Based and Transition-Based DependencyParsers[C]. Proceedings of ACL2008.2008:950–958.
[45] Koo T, Carreras X, Collins M. Simple Semi-supervised Dependency Parsing[C].Proceedings of ACL-08: HLT.2008:595–603.
[46] Chen W, Kazama J, Uchimoto K, et al. Improving Dependency Parsing with Sub-trees from Auto-Parsed Data[C]. Proceedings of the2009Conference on EmpiricalMethods in Natural Language Processing.2009:570–579.
[47] McClosky D, Charniak E, Johnson M. Efective Self-Training for Parsing[C].Proceedings of the Human Language Technology Conference of the NAACL.2006:152–159.
[48] McClosky D, Charniak E, Johnson M. Reranking and Self-Training for ParserAdaptation[C]. Proceedings of the21st International Conference on Computation-al Linguistics and44th Annual Meeting of the Association for Computational Lin-guistics.2006:337–344.
[49] McClosky D, Charniak E, Johnson M. When is Self-Training Efective for Pars-ing?[C]. Proceedings of the22nd International Conference on Computational Lin-guistics (Coling2008).2008:561–568.
[50] Charniak E, Johnson M. Coarse-to-Fine n-Best Parsing and MaxEnt DiscriminativeReranking[C]. Proceedings of ACL-05.2005:173–180.
[51] Chu Y J, Liu T H. On the shortest arborescence of a directed graph[J]. ScienceSinica,1965,14:1396–1400.
[52] McDonald R, Pereira F, Ribarov K, et al. Non-Projective Dependency Parsing us-ing Spanning Tree Algorithms[C]. Proceedings of Human Language TechnologyConference and Conference on Empirical Methods in Natural Language Process-ing.2005:523–530.
[53] Nivre J, Nilsson J. Pseudo-Projective Dependency Parsing[C]. Proceedings of the43rd Annual Meeting of the Association for Computational Linguistics (ACL05).2005:99–106.
[54] Nivre J. Non-Projective Dependency Parsing in Expected Linear Time[C]. Pro-ceedings of the Joint Conference of the47th Annual Meeting of the ACL and the4th International Joint Conference on Natural Language Processing of the AFNLP.2009:351–359.
[55] McClosky D, Charniak E, Johnson M. Automatic Domain Adaptation for Pars-ing[C]. Human Language Technologies: The2010Annual Conference of the NorthAmerican Chapter of the Association for Computational Linguistics.2010:28–36.
[56]段湘煜,赵军,徐波.基于动作建模的中文依存句法分析[J].中文信息学报,2007,21(5).
[57]辛霄,范士喜,王轩,王晓龙.基于最大熵的依存句法分析[J].中文信息学报,2009,23(2).
[58]鉴萍,宗成庆.基于序列标注模型的分层式依存句法分析方法[J].中文信息学报,2010,24(6).
[59]计峰,邱锡鹏.基于序列标注的中文依存句法分析方法[J].计算机应用与软件,2009,26(10).
[60] Duan X, Zhao J,, et al. Probabilistic models for action-based Chinese dependencyparsing[C]. Proceedings of ECML/ECPPKDD.2007.
[61] Titov I, Henderson J. Fast and Robust Multilingual Dependency Parsing witha Generative Latent Variable Model[C]. Proceedings of the CoNLL Shared TaskSession of EMNLP-CoNLL2007.2007:947–951.
[62] Eisner J. Efcient Normal-Form Parsing for Combinatory Categorial Grammar[C].Proceedings of the34th Annual Meeting of the Association for Computational Lin-guistics (ACL).1996:79–86.
[63] Chen W, Kazama J, Tsuruoka Y, et al. Improving Graph-based Dependency Parsingwith Decision History[C]. Coling2010: Posters.2010:126–134.
[64] Zhang H, McDonald R. Generalized Higher-Order Dependency Parsing withCube Pruning[C]. Proceedings of the2012Joint Conference on Empirical Method-s in Natural Language Processing and Computational Natural Language Learning.2012:320–331.
[65] Nunberg G. The linguistics of punctuation[M]. Lecture Notes, no.18. Stanford,CA: Center for the Study of Language and Information,1990.
[66] Collins M. Head-Driven Statistical Models for Natural Language Parsing[C].Computational Linguistics.2003，29:589–637.
[67] White M, Rajkumar R. A More Precise Analysis of Punctuation for Broad-Coverage Surface Realization with CCG[C]. Coling2008: workshop on GrammarEngineering Across Frameworks.2008:17–24.
[68] xun Jin M, Kim M Y, Kim D, et al. Segmentation of Chinese Long SentencesUsing Commas[C]. ACL SIGHAN Workshop2004.2004.
[69] Li X, Zong C, Hu R. A Hierarchical Parsing Approach with Punctuation Processingfor Long Chinese Sentences[C]. IJCNLP2005: Companion Volume.2005.
[70]毛奇,连乐新,周文翠,袁春风.基于标点符号分割的汉语句法分析算法[J].中文信息学报,2007,21(2).
[71] Bohnet B. Efcient Parsing of Syntactic and Semantic Dependency Structures[C].Proceedings of the Thirteenth Conference on Computational Natural LanguageLearning (CoNLL2009): Shared Task.2009:67–72.
[72] Bohnet B. Top Accuracy and Fast Dependency Parsing is not a Contradiction[C].Proceedings of the23rd International Conference on Computational Linguistics(Coling2010).2010:89–97.
[73] Huang L, Chiang D. Better k-best Parsing[C]. Proceedings of IWPT2005.2005.
[74] Gesmundo A, Henderson J, Merlo P, et al. A Latent Variable Model of Syn-chronous Syntactic-Semantic Parsing for Multiple Languages[C]. Proceedings ofCoNLL2009: Shared Task.2009:37–42.
[75] Che W, Li Z, Li Y, et al. Multilingual Dependency-based Syntactic and SemanticParsing[C]. Proceedings of CoNLL2009: Shared Task.2009:49–54.
[76] Xue N, Xia F, Chiou F D, et al. The Penn Chinese TreeBank: Phrase StructureAnnotation of a Large Corpus[C]. Natural Language Engineering.2005，11:207–238.
[77] Zhang Y, Clark S. Joint Word Segmentation and POS Tagging Using a SinglePerceptron[C]. Proceedings of ACL-08: HLT.2008:888–896.
[78] Jiang W, Huang L, Liu Q, et al. A Cascaded Linear Model for Joint ChineseWord Segmentation and Part-of-Speech Tagging[C]. Proceedings of ACL-08: HLT.2008:897–904.
[79] Kruengkrai C, Uchimoto K, Kazama J, et al. An Error-Driven Word-CharacterHybrid Model for Joint Chinese Word Segmentation and POS Tagging[C]. Pro-ceedings of the Joint Conference of the47th Annual Meeting of the ACL and the4th International Joint Conference on Natural Language Processing of the AFNLP.2009:513–521.
[80] Toutanova K, Cherry C. A global model for joint lemmatization and part-of-speechprediction[C]. Proceedings of the Joint Conference of the47th Annual Meeting ofthe ACL and the4th International Joint Conference on Natural Language Process-ing of the AFNLP.2009:486–494.
[81] Cohen S B, Smith N A. Joint Morphological and Syntactic Disambiguation[C].Proceedings of EMNLP-CoNLL2007.2007:208–217.
[82] Goldberg Y, Tsarfaty R. A Single Generative Model for Joint MorphologicalSegmentation and Syntactic Parsing[C]. Proceedings of ACL-08: HLT.2008:371–379.
[83] Finkel J R, Manning C D. Joint Parsing and Named Entity Recognition[C].Proceedings of Human Language Technologies: The2009Annual Conferenceof the North American Chapter of the Association for Computational Linguistics.2009:326–334.
[84] Li J, Zhou G, Ng H T. Joint Syntactic and Semantic Parsing of Chinese[C]. Pro-ceedings of the48th Annual Meeting of the Association for Computational Lin-guistics.2010:1108–1117.
[85] Che W, Liu T. Jointly Modeling WSD and SRL with Markov Logic[C]. Proceed-ings of the23rd International Conference on Computational Linguistics (Coling2010).2010:161–169.
[86] Dyer C. Using a maximum entropy model to build segmentation lattices forMT[C]. Proceedings of Human Language Technologies: The2009Annual Con-ference of the North American Chapter of the Association for Computational Lin-guistics.2009:406–414.
[87] Xiao X, Liu Y, Hwang Y, et al. Joint Tokenization and Translation[C]. Proceedingsof the23rd International Conference on Computational Linguistics (Coling2010).2010:1200–1208.
[88] Liu Y, Liu Q. Joint Parsing and Translation[C]. Proceedings of the23rd Interna-tional Conference on Computational Linguistics (Coling2010).2010:707–715.
[89] McDonald R, Lerman K, Pereira F. Multilingual Dependency Analysis with a Two-Stage Discriminative Parser[C]. Proceedings of the Tenth Conference on Compu-tational Natural Language Learning (CoNLL-X).2006:216–220.
[90] Lee J, Naradowsky J, Smith D A. A Discriminative Model for Joint Morpholog-ical Disambiguation and Dependency Parsing[C]. Proceedings of the49th Annu-al Meeting of the Association for Computational Linguistics: Human LanguageTechnologies.2011:885–894.
[91] Rush A M, Sontag D, Collins M, et al. On Dual Decomposition and Linear Pro-gramming Relaxations for Natural Language Processing[C]. Proceedings of the2010Conference on Empirical Methods in Natural Language Processing.2010:1–11.
[92] Auli M, Lopez A. A Comparison of Loopy Belief Propagation and Dual De-composition for Integrated CCG Supertagging and Parsing[C]. Proceedings of the49th Annual Meeting of the Association for Computational Linguistics: HumanLanguage Technologies.2011:470–480.
[93] Ratnaparkhi A. A Maximum Entropy Model for Part-Of-Speech Tagging[C]. Pro-ceedings of EMNLP1996.1996.
[94] Laferty J, McCallum A, Pereira F. Conditional Random Fields: ProbabilisticModels for Segmenting and Labeling Sequence Data[C]. Proceedings of ICML2001.2001:282–289.
[95] Collins M, Globerson A, Koo T, et al. Exponentiated Gradient Algorithms forConditional Random Fields and Max-Margin Markov Networks[J]. JMLR,2008,9:1775–1822.
[96] Eisner J. Three New Probabilistic Models for Dependency Parsing: An Explo-ration[C]. Proceedings of COLING1996.1996:340–345.
[97] Zhang Y, Clark S. Syntactic processing using the generalized perceptron and beamsearch[J]. Computational Linguistics,2011,37(1):105–151.
[98] Liu X, Zhou M, Zhou X, et al. Joint Inference of Named Entity Recognitionand Normalization for Tweets[C]. Proceedings of the50th Annual Meeting of theAssociation for Computational Linguistics.2012:526–535.
[99] Constant M, Sigogne A, Watrin P. Discriminative Strategies to Integrate Mul-tiword Expression Recognition and Parsing[C]. Proceedings of the50th AnnualMeeting of the Association for Computational Linguistics (Volume1: Long Paper-s).2012:204–212.
[100] Minkov E, Zettlemoyer L. Discriminative Learning for Joint Template Filling[C].Proceedings of the50th Annual Meeting of the Association for Computational Lin-guistics.2012:845–853.
[101] Petrov S, Klein D. Improved Inference for Unlexicalized Parsing[C]. Proceedingsof NAACL2007.2007.
[102] Johansson R, Nugues P. Dependency-based Semantic Role Labeling of Prop-Bank[C]. EMNLP-2008.2008.
[103] Bansal M, Klein D. Web-Scale Features for Full-Scale Parsing[C]. Proceedings ofthe49th Annual Meeting of the Association for Computational Linguistics: HumanLanguage Technologies.2011:693–702.
[104] Zhou G, Zhao J, Liu K, et al. Exploiting Web-Derived Selectional Preference to Im-prove Statistical Dependency Parsing[C]. Proceedings of the49th Annual Meetingof the Association for Computational Linguistics: Human Language Technologies.2011:1556–1565.
[105] Burkett D, Klein D. Two Languages are Better than One (for Syntactic Parsing)[C].Proceedings of the2008Conference on Empirical Methods in Natural LanguageProcessing.2008:877–886.
[106] Chen W, Kazama J, Torisawa K. Bitext Dependency Parsing with Bilingual SubtreeConstraints[C]. Proceedings of the48th Annual Meeting of the Association forComputational Linguistics.2010:21–29.
[107]王跃龙,姬东鸿.汉语树库综述[J].当代语言学,2009,11(1).
[108] Liu T, Ma J, Li S. Building a Dependency Treebank for Improving Chinese Pars-er[C]. Journal of Chinese Language and Computing.2006，16:207–224.
[109] Chen K J, Luo C C, Chang M C, et al. Sinica treebank: Design criteria, rep-resentational issues and implementation[M]. Kluwer Academic Publishers,2003:231–248.
[110]周强.汉语句法树库标注体系[J].中文信息学报,2004,18(4).
[111] Niu Z Y, Wang H, Wu H. Exploiting Heterogeneous Treebanks for Parsing[C]. Pro-ceedings of the Joint Conference of the47th Annual Meeting of the ACL and the4th International Joint Conference on Natural Language Processing of the AFNLP.2009:46–54.
[112] Jiang W, Huang L, Liu Q. Automatic Adaptation of Annotation Standards: ChineseWord Segmentation and POS Tagging–A Case Study[C]. Proceedings of the JointConference of the47th Annual Meeting of the ACL and the4th International JointConference on Natural Language Processing of the AFNLP.2009:522–530.
[113] Smith D A, Eisner J. Parser Adaptation and Projection with Quasi-SynchronousGrammar Features[C]. Proceedings of the2009Conference on Empirical Methodsin Natural Language Processing.2009:822–831.
[114] Wang J N, Chang J S, Su K Y. An Automatic Treebank Conversion Algorithm forCorpus Sharing[C]. Proceedings of the32nd Annual Meeting of the Associationfor Computational Linguistics.1994:248–254.
[115] Collins M, Ramshaw L, Hajic J, et al. A Statistical Parser for Czech[C]. ACL1999.1999:505–512.
[116] Xia F, Palmer M. Converting Dependency Structures to Phrase Structures[C]. InProceedings of HLT2001.2001:1–5.
[117] Xia F, Bhatt R, Rambow O, et al. Towards a Multi-Representational Treebank[C].In Proceedings of the7th International Workshop on Treebanks and Linguistic The-ories.2008.
[118] Li Z, Che W, Liu T. A Study on Constituent-to-Dependency Conversion[J]. Journalof Chinese Information Processing (in Chinese),2008,6(22):14–19.
[119] Martins A F T, Das D, Smith N A, et al. Stacking Dependency Parsers[C].EMNLP’08.2008:157–166.
[120] Smith D, Eisner J. Quasi-Synchronous Grammars: Alignment by Soft Projection ofSyntactic Dependencies[C]. Proceedings on the Workshop on Statistical MachineTranslation.2006:23–30.
[121] Gimpel K, Smith N A. Quasi-Synchronous Phrase Dependency Grammars for Ma-chine Translation[C]. Proceedings of the2011Conference on Empirical Methodsin Natural Language Processing.2011:474–485.
[122] Woodsend K, Lapata M. Learning to Simplify Sentences with Quasi-SynchronousGrammar and Integer Programming[C]. Proceedings of the2011Conference onEmpirical Methods in Natural Language Processing.2011:409–420.
[123] Charniak E. A Maximum-Entropy-Inspired Parser[C]. ANLP’00.2000:132–139.