Stanford NLP解析器是否具有用于语义角色标记的方法? [英] Does the Stanford NLP Parser have methods for semantic role labelling?

问题描述

我正在尝试查找英语句子的语义标签. 我正在使用Stanford NLP解析器.有这个方法吗?

I'm trying to find the semantic labels of english sentences. I am using the Stanford NLP parser. Does it have methods for this?

我正在浏览文档，但是我能找到的最接近的东西是:

I was going through the documentation but the closest thing I could find was:

CoreAnnotations.SemanticWordAnnotation
CoreAnnotations.SemanticTagAnnotation

推荐答案

否，我们目前在CoreNLP中没有语义角色标记(SRL)系统.

No, we currently don't have a semantic role labeling (SRL) system in CoreNLP.

除非您已经有一个明确需要语义角色标签的系统，否则我建议您看一下通用依赖项表示形式.尽管该表示形式主要是一种语法表示形式，但它也编码了很多语义，并且在许多语义任务方面与SRL一样好.

Unless you already have a system that explicitly requires semantic role labels, I would recommend taking a look at the Universal Dependencies representation. Despite the fact that this representation is primarily a syntactic representation, it also encodes a lot semantics and works just as well as SRL for many semantic tasks.

如果该替代方法不适合您，则还存在其他研究小组提供的一些SRL系统，例如UIUC的 SRL系统.

If this alternative does not work for you, then there also exist several SRL systems from other research groups, e.g. the SRL system from UIUC.

这篇关于Stanford NLP解析器是否具有用于语义角色标记的方法?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！