Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

为什么我给的简单句子,大部分都提取不出三元组,比如图中的例子。 #9

Open
XuZhen001 opened this issue Aug 30, 2019 · 11 comments

Comments

@XuZhen001
Copy link

XuZhen001 commented Aug 30, 2019

image

@lemonhu
Copy link
Owner

lemonhu commented Aug 30, 2019

目前只能保证7种DSNF范式,而且这个前提是依存句法正确得到解析。

@XuZhen001
Copy link
Author

这个跟词典有关系吗,resource里面只提供了法律方面的词典,加个通用词典效果会好一点吗?

@XuZhen001
Copy link
Author

image
这个为什么都提不了呢,这个不是标准的“n-v-n”结构吗?谢谢解答

@lemonhu
Copy link
Owner

lemonhu commented Aug 30, 2019

添加用户词典,只是有助于分词这一步。

建立关系的候选实体集合为{'ns', 'ni', 'nh', 'nz', 'j'},可见entity_combine.py

@XuZhen001
Copy link
Author

好的,非常感谢,那么你觉得怎么改进呢,比如可以识别更多句法什么的,要大致怎么改呢?

@lemonhu
Copy link
Owner

lemonhu commented Aug 30, 2019

这份工作的贡献就是基于依存句法制定抽取范式,我认为可以从以下两个方面提高:

  1. 制定更多的抽取范式,而范式的制定需要较深的语言学知识,当然规则的复杂性可能会随之增加。
  2. 实际测试中,长句子的依存句法解析依然比较困难,可以尝试解决长句子的依存解析不准的问题。

@XuZhen001
Copy link
Author

好的,非常感谢

@sssdjj
Copy link

sssdjj commented Oct 31, 2019

用自己训练的命名实体识别+结巴分词能优化很多大准确率,谢谢作者的贡献

@Bernard-Yang
Copy link

用自己训练的命名实体识别+结巴分词能优化很多大准确率,谢谢作者的贡献

您好,可以请教下具体怎么改进的吗

@tianyunzhe
Copy link

用自己训练的命名实体识别+结巴分词能优化很多大准确率,谢谢作者的贡献

是嘛,我目前从分词,词性标注,命名实体识别,依存分析还是清一色的LTP。

@onion0925
Copy link

用自己训练的命名实体识别+结巴分词能优化很多大准确率,谢谢作者的贡献

您好,想问一下怎样可以训练自己的数据。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants