arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1609.09247cs.CLcs.LG
收藏

使用部分注释训练依赖分析器

Training Dependency Parsers with Partial Annotation

Soochow University(苏州大学)

Zhenghua Li, Yue Zhang, Jiayuan Chao, Min Zhang

摘要:

最近,研究如何获得部分注释数据以进行模型监督的研究激增。然而,如何训练具有部分标注(PA)的统计模型还缺乏系统的研究。本文以依存句法分析为例,描述并比较了三种主流依赖解析器的两种直接方法。第一种方法是以前提出的直接训练一个对数线性图为基础的分析器(LLGPar)与PA基于森林的目标。这项工作首次提出了第二种方法,直接训练线性图为基础的解析器(LGPar)和线性转换为基础的解析器(LTPar)与PA基于约束解码的想法。我们在Penn Treebank上进行了广泛的实验,在三种不同的设置下模拟PA,即,随机依赖关系、最不确定依赖关系以及具有来自三个解析器的不同输出的依赖关系。结果表明,LLGPar是最有效的学习PA和LTPar落后于基于图形的同行大幅度。此外,LGPar和LTPar可以通过使用LLGPar将PA完成为完整注释(FA)来实现最佳性能。

英文摘要:

Recently, these has been a surge on studying how to obtain partially annotated data for model supervision. However, there still lacks a systematic study on how to train statistical models with partial annotation (PA). Taking dependency parsing as our case study, this paper describes and compares two straightforward approaches for three mainstream dependency parsers. The first approach is previously proposed to directly train a log-linear graph-based parser (LLGPar) with PA based on a forest-based objective. This work for the first time proposes the second approach to directly training a linear graph-based parse (LGPar) and a linear transition-based parser (LTPar) with PA based on the idea of constrained decoding. We conduct extensive experiments on Penn Treebank under three different settings for simulating PA, i.e., random dependencies, most uncertain dependencies, and dependencies with divergent outputs from the three parsers. The results show that LLGPar is most effective in learning from PA and LTPar lags behind the graph-based counterparts by large margin. Moreover, LGPar and LTPar can achieve best performance by using LLGPar to complete PA into full annotation (FA).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑