Transition-Based Dependency Parsing Exploiting Supertags

被引:8
作者
Ouchi, Hiroki [1 ]
Duh, Kevin [1 ]
Shindo, Hiroyuki [1 ]
Matsumoto, Yuji [1 ]
机构
[1] Nara Inst Sci & Technol, Dept Informat Sci, Ikoma, Nara 6300101, Japan
关键词
Dependency parsing; multilingual dependency parsing; supertags; transition-based dependency parsing;
D O I
10.1109/TASLP.2016.2598310
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
Lexical information, including surface word form and part-of-speech (POS) information, plays a crucial role when predicting ambiguous dependency relationships in dependency parsing. However, for resolving dependency ambiguities, surface word information may be too sparse, while POS information may be too coarse. Supertags, which are lexical templates that represent rich syntactic information, have been shown to provide effective features at an intermediate level on the coarse-to-fine scale. In this work, we present a supertag design framework that allows us to instantiate various supertag sets based on the dependency structures. Using this framework, we instantiate various supertag sets and utilize them as features in transition-based dependency parsing systems. Performing experiments on the Penn Treebank and Universal Dependencies data sets, we show that our supertags are effective for transition-based parsers in multilingual parsing as well as English parsing. The comparison of the results of the different supertag sets shows that it is crucial to incorporate the head directionality, head labels, and dependent possession information in supertags to improve the parser performance.
引用
收藏
页码:2059 / 2068
页数:10
相关论文
共 30 条
[1]  
Ambati BharatRam., 2014, EACL, P159
[2]  
[Anonymous], 2003, P 8 INT C PARSING TE
[3]  
[Anonymous], 2003, P 8 INT C PARS TECHN
[4]  
[Anonymous], 2012, P 2012 JOINT C EMP M
[5]  
[Anonymous], 2011, P 49 ANN M ASS COMP
[6]  
[Anonymous], 2009, Dependency Parsing
[7]  
[Anonymous], 1993, COMPUT LINGUIST, DOI DOI 10.21236/ADA273556
[8]  
[Anonymous], 2006, PROC 5 INT C LANGUAG
[9]  
[Anonymous], 2014, P 18 C EMP METH NAT, DOI DOI 10.3115/V1/D14-1082
[10]  
[Anonymous], 2008, Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing