自然语言处理基础技术之词性标注实战

一些常用python工具包的词性标注demo，供小白入门练手找感觉~~

jieba词性标注（part of specch)

安装：pip install jieba

国内源安装更快：pip install jieba -i https://pypi.tuna.tsinghua.edu.cn/simple

先导包：jieba.posseg.dt 为默认词性标注分词器

标注句子分词后每个词的词性，采用和 ictclas 兼容的标记法。

jieba貌似不能处理英文，后面会介绍处理英文的

import jieba.posseg as pseg

words = pseg.cut("我爱自然语言处理技术！")

for word, pos in words:

print(word, pos)

我 r 爱 v 自然语言 l 处理 v 技术 n ！ x

SnowNLP词性标注

安装：pip install snownlp

国内源安装：pip install snownlp -i https://pypi.tuna.tsinghua.edu.cn/simple

使用snownlp进行词性标注

from snownlp import SnowNLP

model = SnowNLP(u'我爱自然语言处理技术！')

for word, pos in model.tags:

print(word, pos)

我 r 爱 v 自然 n 语言 n 处理 vn 技术 n ！ w

THULAC词性标注

安装：pip install thulac

国内源安装：pip install thulac -i https://pypi.tuna.tsinghua.edu.cn/simple

使用thulac进行词性标注

import thulac thulac_model = thulac.thulac()

wordseg = thulac_model.cut("我爱自然语言处理技术！")

print(wordseg)

Model loaded succeed

[['我', 'r'], ['爱', 'v'], ['自然', 'n'], ['语言', 'n'], ['处理', 'v'], ['技术', 'n'], ['！', 'w']]

Stanford CoreNLP分词

安装：pip install stanfordcorenlp

国内源安装：pip install stanfordcorenlp -i https://pypi.tuna.tsinghua.edu.cn/simple

使用stanfordcorenlp进行词性标注

同时支持英文和中文的词性标注

from stanfordcorenlp import StanfordCoreNLP

zh_model = StanfordCoreNLP(r'stanford-corenlp-full-2018-02-27',)

s_zh = '我爱自然语言处理技术！'

word_pos_zh = zh_model.pos_tag(s_zh)

print(word_pos_zh)

[('我爱', 'NN'), ('自然', 'AD'), ('语言', 'NN'), ('处理', 'VV'), ('技术', 'NN'), ('！', 'PU')]

eng_model = StanfordCoreNLP(r'stanford-corenlp-full-2018-02-27')

s_eng = 'I love natural language processing technology!'

word_pos_eng = eng_model.pos_tag(s_eng)

print(word_pos_eng)

[('I', 'PRP'), ('love', 'VBP'), ('natural', 'JJ'), ('language', 'NN'), ('processing', 'NN'), ('technology', 'NN'), ('!', '.')]

Hanlp词性标注

安装：pip install pyhanlp

国内源安装：pip install pyhanlp -i https://pypi.tuna.tsinghua.edu.cn/simple

使用pyhanlp进行词性标注

from pyhanlp import *

s = '我爱自然语言处理技术！'

word_seg = HanLP.segment(s)

for term in word_seg:

print(term.word, term.nature)

我 rr 爱 v 自然语言处理 nz 技术 n ！ w

NLTK词性标注

安装：pip install nltk

国内源安装：pip install nltk -i https://pypi.tuna.tsinghua.edu.cn/simple

nltk只能处理英文

import nltk

s = 'I love natural language processing technology!'

s = nltk.word_tokenize(s)

s_pos = nltk.pos_tag(s)

print(s_pos)

[('I', 'PRP'), ('love', 'VBP'), ('natural', 'JJ'), ('language', 'NN'), ('processing', 'NN'), ('technology', 'NN'), ('!', '.')]

spaCy词性标注

安装：pip install spaCy

国内源安装：pip install spaCy -i https://pypi.tuna.tsinghua.edu.cn/simple

下载不了模型，需要python -m spacy download en。

The easiest solution is to re-run the command as admin(意思是用用户管理权限打开CMD下载即可)

import spacy eng_model = spacy.load('en')

s = 'I love natural language processing technology!'

词性标注

s_token = eng_model(s)

for token in s_token:

print(token, token.pos_, token.pos)

I PRON 94 love VERB 99 natural ADJ 83 language NOUN 91 processing NOUN 91 technology NOUN 91 ! PUNCT 96

另外，代码我已经上传github：https://github.com/yuquanle/StudyForNLP/blob/master/NLPbasic/POS.ipynb

更多个人笔记请关注：

公众号：StudyForAI（小白人工智能入门学习）

知乎专栏：https://www.zhihu.com/people/yuquanle/columns