自然语言处理基础技术工具篇之NLTK

NLTK简介

• NLTK被称为“使用Python进行计算语言学教学和工作的绝佳工具”。 它为50多种语料库和词汇资源（如WordNet）提供了易于使用的界面，还提供了一套用于分类，标记化，词干化，标记，解析和语义推理的文本处理库。接下来然我们一起来实战学习一波~~
• Github地址：https://github.com/nltk/nltk
• 官方文档：http://www.nltk.org/

---

NLTK

安装：pip install nltk

---

1.Tokenize

import nltk

sentence = 'I love natural language processing!'

tokens = nltk.word_tokenize(sentence)
print(tokens)

['I', 'love', 'natural', 'language', 'processing', '!']

2.词性标注

tagged = nltk.pos_tag(tokens)
print(tagged)

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

3.命名实体识别

• 下载模型：nltk.download(‘maxent_ne_chunker’)

nltk.download('maxent_ne_chunker')

[nltk_data] Downloading package maxent_ne_chunker to
[nltk_data]     C:\Users\yuquanle\AppData\Roaming\nltk_data...
[nltk_data]   Unzipping chunkers\maxent_ne_chunker.zip.





True

nltk.download('words')

[nltk_data] Downloading package words to
[nltk_data]     C:\Users\yuquanle\AppData\Roaming\nltk_data...
[nltk_data]   Unzipping corpora\words.zip.





True

entities = nltk.chunk.ne_chunk(tagged)
print(entities)

(S I/PRP love/VBP natural/JJ language/NN processing/NN !/.)

4.下载语料库

• 例如：下载brown
• 更多语料库：http://www.nltk.org/howto/corpus.html

nltk.download('brown')

[nltk_data] Downloading package brown to
[nltk_data]     C:\Users\yuquanle\AppData\Roaming\nltk_data...
[nltk_data]   Package brown is already up-to-date!





True

from nltk.corpus import brown
brown.words()

['The', 'Fulton', 'County', 'Grand', 'Jury', 'said', ...]

5.度量

• percision：正确率
• recall：召回率
• f_measure

from nltk.metrics import precision, recall, f_measure
reference = 'DET NN VB DET JJ NN NN IN DET NN'.split()
test    = 'DET VB VB DET NN NN NN IN DET NN'.split()
reference_set = set(reference)
test_set = set(test)

print("precision:" + str(precision(reference_set, test_set)))
print("recall:" + str(recall(reference_set, test_set)))
print("f_measure:" + str(f_measure(reference_set, test_set)))

precision:1.0
recall:0.8
f_measure:0.8888888888888888

6.词干提取(Stemmers)

• Porter stemmer

from nltk.stem.porter import *
# 创建词干提取器
stemmer = PorterStemmer()
plurals = ['caresses', 'flies', 'dies', 'mules', 'denied']
singles = [stemmer.stem(plural) for plural in plurals]
print(' '.join(singles))

caress fli die mule deni

• Snowball stemmer

from nltk.stem.snowball import SnowballStemmer
print(" ".join(SnowballStemmer.languages))

arabic danish dutch english finnish french german hungarian italian norwegian porter portuguese romanian russian spanish swedish

# 指定语言
stemmer = SnowballStemmer("english")
print(stemmer.stem("running"))

run

7.SentiWordNet接口

• 下载sentiwordnet词典

import nltk
nltk.download('sentiwordnet')

[nltk_data] Downloading package sentiwordnet to
[nltk_data]     C:\Users\yuquanle\AppData\Roaming\nltk_data...
[nltk_data]   Unzipping corpora\sentiwordnet.zip.





True

• SentiSynsets: synsets(同义词集)的情感值

from nltk.corpus import sentiwordnet as swn

breakdown = swn.senti_synset('breakdown.n.03')
print(breakdown)
print(breakdown.pos_score())
print(breakdown.neg_score())
print(breakdown.obj_score())

<breakdown.n.03: PosScore=0.0 NegScore=0.25>
0.0
0.25
0.75

• Lookup(查看)

print(list(swn.senti_synsets('slow')))

[SentiSynset('decelerate.v.01'), SentiSynset('slow.v.02'), SentiSynset('slow.v.03'), SentiSynset('slow.a.01'), SentiSynset('slow.a.02'), SentiSynset('dense.s.04'), SentiSynset('slow.a.04'), SentiSynset('boring.s.01'), SentiSynset('dull.s.08'), SentiSynset('slowly.r.01'), SentiSynset('behind.r.03')]

happy = swn.senti_synsets('happy', 'a')
print(list(happy))

[SentiSynset('happy.a.01'), SentiSynset('felicitous.s.02'), SentiSynset('glad.s.02'), SentiSynset('happy.s.04')]

---

更多资料

• 更多用法：http://www.nltk.org/howto/index.html
• Github地址:https://github.com/nltk/nltk

欢迎关注【AI小白入门】，这里分享Python、机器学习、深度学习、自然语言处理、人工智能等技术，关注前沿技术，求职经验等，陪有梦想的你一起成长。