Stanford CoreNLP错误的共同引用解决方案 [英] Stanford CoreNLP wrong coreference resolution

问题描述

我仍在使用斯坦福大学的CoreNLP，并且在对共指分辨率进行非常琐碎的测试时遇到了奇怪的结果.

I am still playing with Stanford's CoreNLP and I am encountering strange results on a very trivial test of Coreference resolution.

给出两个句子:

酒店有一个大浴室.很干净.

The hotel had a big bathroom. It was very clean.

我希望句子2中的它"与句子1中的浴室"或至少大浴室"同时使用.

I would expect "It" in sentence 2 to be coreferenced by "bathroom" or at least "a big bathroom" of sentence 1.

不幸的是，它指向酒店"，我认为这是错误的.

Unfortunately it point to "The hotel" which in my opinion is wrong.

有没有办法解决这个问题?我需要训练什么东西还是应该立即使用?

Is there a way to solve this problem ? Do I need to train anything or is it supposed to work out of the box ?

    Annotation a = getPipeline().getAnnotation("The hotel had a big bathroom. It was very clean.");

    System.out.println(a.get(CorefChainAnnotation.class));

输出:

{1 = CHAIN1- [句子1中的旅馆"，句子2中的它"，2 = CHAIN2- [句子1中的大浴室"]}

{1=CHAIN1-["The hotel" in sentence 1, "It" in sentence 2], 2=CHAIN2-["a big bathroom" in sentence 1]}

非常感谢您的帮助.

推荐答案

与AI中的许多组件一样，斯坦福共指系统仅在一定程度上是正确的.在共参照的情况下，此精度实际上相对较低(在0-100范围内的标准基准下约为60).为了说明该问题的难度，请考虑以下具有相似的共同参照判断的明显相似的句子:

Like many components in AI, the Stanford coreference system is only correct to a certain accuracy. In the case of coreference this accuracy is actually relatively low (~60 on standard benchmarks in a 0-100 range). To illustrate the difficulty of the problem, consider the following apparently similar sentence with a different coreference judgment:

酒店有一个大浴缸.这非常昂贵.

The hotel had a big bathtub. It was very expensive.

这篇关于Stanford CoreNLP错误的共同引用解决方案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！