当前位置:网站首页>Named entity recognition in natural language processing: tanford core LP ner (1)

Named entity recognition in natural language processing: tanford core LP ner (1)

2020-11-06 01:28:24 Elementary school students in IT field

Reprint please indicate the source :https://blog.csdn.net/HHTNAN

brief introduction

CoreNLP The project is Stanford Developed a set of open source NLP System . Include tokenize, pos , parse And so on , And SpaCy similar .SpaCy Claims to be the fastest NLP System , And provide ready-made python Interface , But the disadvantage is that it does not support Chinese processing at present , CoreNLP The Chinese model is included , It can be directly used to process Chinese , but CoreNLP Use Java Development ,python It's a little bit cumbersome to call .

Stanford CoreNLP It is a powerful natural language processing tool , Many models are trained based on deep learning .

First attach a link to its official website :

https://stanfordnlp.github.io/CoreNLP/index.html
https://nlp.stanford.edu/nlp/javadoc/javanlp/
https://github.com/stanfordnlp/CoreNLP

install Installation

windows 10 Environmental Science
Installation dependency
1. First you need to configure JDK, install JDK 1.8 And above ..
2. Later on https://stanfordnlp.github.io/CoreNLP/history.html Download the corresponding jar package .
Decompress the package to get the directory , Then the language of jar Put the package in this directory .
3. download Stanford CoreNLP file :http://stanfordnlp.github.io/CoreNLP/download.html
 Insert picture description here
4. Download the Chinese model jar package ( Be sure to download this file , Otherwise, it is handled in English by default ).
 Insert picture description here
5. Next py install stanfordcorenlp
 Insert picture description here
6. Unzip configuration
When the download is complete, the two files add up 1G+ When the download is complete, the two files add up 1G+
 Insert picture description here
Take the decompressed Stanford CoreNLP Folder downloaded Stanford-chinese-corenlp-2018—models.jar Put it in the same directory ( Be careful : Be sure to be in the same directory , Otherwise, the execution will report an error )
 Insert picture description here
7. stay Python Reference model in , Execute the following statement :

from stanfordcorenlp import StanfordCoreNLP
nlp=StanfordCoreNLP(r’D:\D:\stanford_nlp\stanford-corenlp-full-2018-10-05’,lang=‘zh’)

application

#encoding="utf-8"
from stanfordcorenlp import StanfordCoreNLP
import os

if os.path.exists('D:\\stanford_nlp\\stanford-corenlp-full-2018-10-05'):
    print("corenlp exists")
else:
    print("corenlp not exists")
nlp=StanfordCoreNLP('D:\\stanford_nlp\\stanford-corenlp-full-2018-10-05',lang='zh')
sentence = ' Wang Ming is a graduate student of Tsinghua University '
print(nlp.ner(sentence))

Output :
corenlp exists

[(‘ Wang Ming ’, ‘PERSON’), (‘ yes ’, ‘O’), (‘ tsinghua ’, ‘ORGANIZATION’), (‘ university ’, ‘ORGANIZATION’), (‘ Of ’, ‘O’), (‘ One ’, ‘NUMBER’), (‘ individual ’, ‘O’), (‘ Graduate student ’, ‘O’)]

3、 ... and 、 See part of speech tagging
Access in a browser :http://localhost:9000/
 Insert picture description here

 WeChat ID

Reprint please indicate the source :https://blog.csdn.net/HHTNAN

版权声明
本文为[Elementary school students in IT field]所创,转载请带上原文链接,感谢