Predicate Argument Structure Analysis using Partially Annotated Corpora
Predicate Argument Structure Analysis using Partially Annotated Corpora
复制标题
DOI：
--
复制
发表时间：
2013-10
期刊：
影响因子：
--
通讯作者：
Koichiro Yoshino;Shinsuke Mori;Tatsuya Kawahara
中科院分区：
文献类型：
--
作者：
Koichiro Yoshino;Shinsuke Mori;Tatsuya Kawahara
我们提出了一种新的可以从部分标注语料库中训练的谓词参数结构分析方案。为了允许部分注释，这个语义角色标签器不需要worddependencyinformation。部分标注的优点是允许对训练数据进行平滑的领域自适应，提高了对各种领域的适应性。
We present a novel scheme of predicate argument structure analysis that can be trained from partially annotated corpora. In order to allow partial annotation, this semantic role labeler does not require worddependencyinformation. Theadvantage of partial annotation is that it allows for smooth domain adaptation of training data and improves the adaptability to a variety of domains.