Structure-guided supertagger learning

Structure-guided supertagger learning
复制标题

结构引导的超级标记学习

DOI:
10.1017/s1351324912000034
复制
发表时间:
2012
影响因子:
2.5
通讯作者:
Jun'ichi Tsujii
Jun'ichi Tsujii
中科院分区:
计算机科学3区
文献类型:
--
作者:
Yao-zhong Zhang;Takuya Matsuzaki;Jun'ichi Tsujii

文献摘要

参考文献

相似文献

正如本文所述,我们专门研究了超级标记任务的结构学习问题。超级标记是一项为句子中的每个单词分配最可能的词汇条目的任务。超级标记器对于词汇化语法解析器极其重要,因为准确的超级标记器可以大大减少下游解析器中的词汇歧义。超级标记比传统的序列标记任务(例如词性标记)更具挑战性。首先,超级标签数量众多。超级标签是词汇化语法中定义的词汇条目,由丰富的句法/语义信息组成。其次,超级标签之间的关系更加复杂。正确的超级标签分配应该与句子中的其他超级标签分配兼容以构造解析树。序列标记模型中常用的相邻标签特征(例如,一阶边缘特征)对于超级标记任务来说过于粗糙。远程信息对于超级标记任务极其重要。提出了两种在超级标记者训练阶段考虑远程信息的方法。具体来说,我们提出了一种依赖项通知的超级标记器,以使用从依赖项解析器派生的单词到单词的依赖项,并生成远程特征作为训练中的软约束。在森林引导的超级标记器中,我们限制分类器在满足语法的空间中学习,并使用 CFG 过滤器对模型参数的更新施加语法约束。实验表明,所提出的结构引导超级标记器的性能明显优于基线超级标记器。基于改进的超级标记器,最终解析器的F-score也得到了提高。在移位归约 HPSG 解析器中使用森林引导超级标记器,我们实现了 89.31% F 分数的竞争性解析性能,并且解析速度比最先进的 HPSG 解析器更高。
As described in this paper, we specifically examine the structural learning problem of a supertagging task. Supertagging is a task to assign the most probable lexical entry to each word in a sentence. A supertagger is extremely important for a lexicalized grammar parser because an accurate supertagger can greatly reduce lexical ambiguity in downstream parser. Supertagging is more challenging than conventional sequence labeling tasks (e.g., part-of-speech tagging). First, the supertags are numerous. Supertags are the lexical entries defined in a lexicalized grammar, which consists of rich syntactic/semantic information. Second, the inter-supertag relation is more complex. A proper supertag assignment is expected to be compatible with other supertag assignments in a sentence to construct a parse tree. Commonly used adjacent label features (e.g., first-order edge feature) in a sequence labeling model are too rough for the supertagging task. Long-range information is extremely important for the supertagging task. Two approaches to consider long-range information in a supertagger's training stage are proposed. Specifically, we propose a dependency-informed supertagger to use word-to-word dependency derived from a dependency parser and generate long-range features as soft constraints in the training. In the forest-guided supertagger, we constrain the classifier to learn in a grammar-satisfying space and use a CFG filter to impose grammar constraints for the update of model parameters. The experiments show that the proposed structure-guided supertaggers perform significantly better than the baseline supertaggers. Based on the improved supertaggers, the F-score of the final parser is also improved. Using the forest-guided supertagger in a shift-reduce HPSG parser, we achieved a competitive parsing performance of 89.31% F-score with higher parsing speed than that of a state-of-the-art HPSG parser.
DOI: --
发表时间: 2007-01
期刊: --
影响因子: --
作者:
Takuya Matsuzaki;Yusuke Miyao;Junichi Tsujii
通讯作者: Takuya Matsuzaki;Yusuke Miyao;Junichi Tsujii
具有 CFG 过滤功能的 HPSG 解析器
DOI: 10.1017/s1351324900002412
发表时间: 2000
影响因子: 2.5
作者:
Kentaro Torisawa;K. Nishida;Yusuke Miyao;Junichi Tsujii
通讯作者: Junichi Tsujii
基于 SNoW 的超级标记器及其在 NP 分块中的应用
DOI: 10.3115/1075096.1075160
发表时间: 2003
影响因子: --
作者:
Libin Shen;A. Joshi
通讯作者: A. Joshi
使用限制来扩展基于复杂特征的形式主义的解析算法
DOI: 10.3115/981210.981228
发表时间: 1985
期刊: --
影响因子: --
作者:
Stuart M. Shieber
通讯作者: Stuart M. Shieber
DOI: 10.3115/1219840.1219851
发表时间: 2005-06
期刊: --
影响因子: --
作者:
Yusuke Miyao;Junichi Tsujii
通讯作者: Yusuke Miyao;Junichi Tsujii