NLTK基础

Python上著名的⾃然语⾔处理库

• ⾃带语料库，词性分类库
• ⾃带分类，分词，等等功能
• 强⼤的社区⽀持
• 还有N多的简单版wrapper

安装语料库

# 方式一
import nltk
nltk.download()

showing info https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/index.xml

若下载速度慢或因其他原因下载失败

官方下载地址 http://www.nltk.org/nltk_data/¶

githup下载地址 https://github.com/nltk/nltk_data

• 下载packages文件，重命名为nltk_data

from nltk import data
data.path.append('D:/python3.6/nltk_data')

功能一览表

下载语料库

# 请下载
nltk.download('brown')

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

nltk自带语料库

# nltk自带语料库
from nltk.corpus import brown
brown.categories()
['adventure',
 'belles_lettres',
 'editorial',
 'fiction',
 'government',
 'hobbies',
 'humor',
 'learned',
 'lore',
 'mystery',
 'news',
 'religion',
 'reviews',
 'romance',
 'science_fiction']
brown.readme()   # 语料信息描述

print(brown.words()[:10])    # 单词
print(len(brown.words()))

['The', 'Fulton', 'County', 'Grand', 'Jury', 'said', 'Friday', 'an', 'investigation', 'of']
1161192

print(brown.sents()[:10])    # 句子
print(brown.sents().__len__())
[['The', 'Fulton', 'County', 'Grand', 'Jury', 'said', 'Friday', 'an', 'investigation', 'of', "Atlanta's", 'recent', 'primary', 'election', 'produced', '``', 'no', 'evidence', "''", 'that', 'any', 'irregularities', 'took', 'place', '.'], ['The', 'jury', 'further', 'said', 'in', 'term-end', 'presentments', 'that', 'the', 'City', 'Executive', 'Committee', ',', 'which', 'had', 'over-all', 'charge', 'of', 'the', 'election', ',', '``', 'deserves', 'the', 'praise', 'and', 'thanks', 'of', 'the', 'City', 'of', 'Atlanta', "''", 'for', 'the', 'manner', 'in', 'which', 'the', 'election', 'was', 'conducted', '.'], ['The', 'September-October', 'term', 'jury', 'had', 'been', 'charged', 'by', 'Fulton', 'Superior', 'Court', 'Judge', 'Durwood', 'Pye', 'to', 'investigate', 'reports', 'of', 'possible', '``', 'irregularities', "''", 'in', 'the', 'hard-fought', 'primary', 'which', 'was', 'won', 'by', 'Mayor-nominate', 'Ivan', 'Allen', 'Jr.', '.'], ['``', 'Only', 'a', 'relative', 'handful', 'of', 'such', 'reports', 'was', 'received', "''", ',', 'the', 'jury', 'said', ',', '``', 'considering', 'the', 'widespread', 'interest', 'in', 'the', 'election', ',', 'the', 'number', 'of', 'voters', 'and', 'the', 'size', 'of', 'this', 'city', "''", '.'], ['The', 'jury', 'said', 'it', 'did', 'find', 'that', 'many', 'of', "Georgia's", 'registration', 'and', 'election', 'laws', '``', 'are', 'outmoded', 'or', 'inadequate', 'and', 'often', 'ambiguous', "''", '.'], ['It', 'recommended', 'that', 'Fulton', 'legislators', 'act', '``', 'to', 'have', 'these', 'laws', 'studied', 'and', 'revised', 'to', 'the', 'end', 'of', 'modernizing', 'and', 'improving', 'them', "''", '.'], ['The', 'grand', 'jury', 'commented', 'on', 'a', 'number', 'of', 'other', 'topics', ',', 'among', 'them', 'the', 'Atlanta', 'and', 'Fulton', 'County', 'purchasing', 'departments', 'which', 'it', 'said', '``', 'are', 'well', 'operated', 'and', 'follow', 'generally', 'accepted', 'practices', 'which', 'inure', 'to', 'the', 'best', 'interest', 'of', 'both', 'governments', "''", '.'], ['Merger', 'proposed'], ['However', ',', 'the', 'jury', 'said', 'it', 'believes', '``', 'these', 'two', 'offices', 'should', 'be', 'combined', 'to', 'achieve', 'greater', 'efficiency', 'and', 'reduce', 'the', 'cost', 'of', 'administration', "''", '.'], ['The', 'City', 'Purchasing', 'Department', ',', 'the', 'jury', 'said', ',', '``', 'is', 'lacking', 'in', 'experienced', 'clerical', 'personnel', 'as', 'a', 'result', 'of', 'city', 'personnel', 'policies', "''", '.']]
57340

print(brown.tagged_words()[:10])    # 词性标注
print(brown.tagged_words().__len__())
[('The', 'AT'), ('Fulton', 'NP-TL'), ('County', 'NN-TL'), ('Grand', 'JJ-TL'), ('Jury', 'NN-TL'), ('said', 'VBD'), ('Friday', 'NR'), ('an', 'AT'), ('investigation', 'NN'), ('of', 'IN')]
1161192

二、文本处理流程

• 1.preprocess

• 2.tokenize

• 3.stopwords

• 4....

• 5.make features

• 6.machine learning

　

一、Tokenize

　　把句子拆成有意义的小部件

import nltk
sentence = 'Never underestimate the heart of a champion '
tokens = nltk.word_tokenize(sentence)
tokens
['Never', 'underestimate', 'the', 'heart', 'of', 'a', 'champion']

　　中文分词

import jieba
seg_list = jieba.cut("我来到北京清华⼤学", cut_all=True)
print("全模式:", "/ ".join(seg_list))  # 全模式
seg_list = jieba.cut("我来到北京清华⼤学", cut_all=False)
print("精确模式:", "/ ".join(seg_list))  # 精确模式
seg_list = jieba.cut("他来到了⽹易杭研⼤厦") # 默认是精确模式
print('新词识别:',", ".join(seg_list))
seg_list = jieba.cut_for_search("⼩明硕⼠毕业于中国科学院计算所，后在⽇本京都⼤学深造")
print('搜索引擎模式：',','.join(seg_list))

全模式: 我/ 来到/ 北京/ 清华/ / / 学
精确模式: 我/ 来到/ 北京/ 清华/ ⼤/ 学
新词识别: 他, 来到, 了, ⽹, 易, 杭研, ⼤, 厦
搜索引擎模式： ⼩,明硕,⼠,毕业,于,中国,科学,学院,科学院,中国科学院,计算,计算所,，,后,在,⽇,本,京都,⼤,学,深造　　

社交网络语言的分词

例子

# 社交网络语言的tokenize
from nltk.tokenize import word_tokenize

tweet = 'RT @angelababy: love you baby! :D http://ah.love #168cm'
print(word_tokenize(tweet))
['RT', '@', 'angelababy', ':', 'love', 'you', 'baby', '!', ':', 'D', 'http', ':', '//ah.love', '#', '168cm']

　解决方法：正则表达式过滤

import re
emoticons_str = r"""
    (?:
        [:=;] # 眼睛
        [oO\-]? # ⿐⼦
        [D\)\]\(\]/\\OpP] # 嘴
    )"""
regex_str = [
    emoticons_str,
    r'<[^>]+>', # HTML tags
    r'(?:@[\w_]+)', # @某⼈
    r"(?:\#+[\w_]+[\w\'_\-]*[\w_]+)", # 话题标签
    r'http[s]?://(?:[a-z]|[0-9]|[$-_@.&+]|[!*\(\),]|(?:%[0-9a-f][0-9a-f]))+',
                   # URLs
    r'(?:(?:\d+,?)+(?:\.?\d+)?)', # 数字
    r"(?:[a-z][a-z'\-_]+[a-z])", # 含有 - 和 ‘ 的单词
    r'(?:[\w_]+)', # 其他
    r'(?:\S)' # 其他
]

正则表达式对照表

http://www.regexlab.com/zh/regref.htm¶

tokens_re = re.compile(r'('+'|'.join(regex_str)+')', re.VERBOSE | re.IGNORECASE)
emoticon_re = re.compile(r'^'+emoticons_str+'$', re.VERBOSE | re.IGNORECASE)

def tokenize(s):
    return tokens_re.findall(s)

def preprocess(s, lowercase=False):
    tokens = tokenize(s)
    if lowercase:
        tokens = [token if emoticon_re.search(token) else token.lower() for token in
tokens]
    return tokens

tweet = 'RT @angelababy: love you baby! :D http://ah.love #168cm'
print(preprocess(tweet))

['RT', '@angelababy', ':', 'love', 'you', 'baby', '!', ':D', 'http://ah.love', '#168cm']

二、词型归一化

纷繁复杂的词形

• Inflection变化: walk => walking => walked
• 影响词性
• derivation 引申: nation (noun) => national (adjective) => nationalize (verb)
• 不影响词性

Stemming 词⼲提取：⼀般来说，就是把不影响词性的inflection的⼩尾巴砍掉

• walking 砍ing = walk
• walked 砍ed = walk
• Lemmatization 词形归⼀：把各种类型的词的变形，都归为⼀个形式
• went 归⼀ = go
• are 归⼀ = be

1. 词干提取

from nltk.stem.porter import PorterStemmer
porter_stemmer = PorterStemmer()
print(porter_stemmer.stem('maximum'))
print(porter_stemmer.stem('presumably'))
print(porter_stemmer.stem('multiply'))
print(porter_stemmer.stem('provision'))

maximum
presum
multipli
provis

2. 词型归一　　

from nltk.stem import WordNetLemmatizer
wordnet_lemmatizer = WordNetLemmatizer()
print(wordnet_lemmatizer.lemmatize('dogs'))
print(wordnet_lemmatizer.lemmatize('churches'))
print(wordnet_lemmatizer.lemmatize('aardwolves'))
print(wordnet_lemmatizer.lemmatize('abaci'))
print(wordnet_lemmatizer.lemmatize('hardrock'))

dog
church
aardwolf
abacus
hardrock

　没有pos tag,，默认是nn

# ⽊有POS Tag，默认是NN 名词
wordnet_lemmatizer.lemmatize('are')
'are'
wordnet_lemmatizer.lemmatize('is')
'is'

　词性标注

　　方式一：手动标注

# 加上POS Tag
print(wordnet_lemmatizer.lemmatize('is', pos='v'))
print(wordnet_lemmatizer.lemmatize('are', pos='v'))
'be'
'be'

　　方式二：

import nltk
text = nltk.word_tokenize('what does the fox say')
print(text)
print(nltk.pos_tag(text))
['what', 'does', 'the', 'fox', 'say']
[('what', 'WDT'), ('does', 'VBZ'), ('the', 'DT'), ('fox', 'NNS'), ('say', 'VBP')]

　词性关系表

三、Stopwords

⼀千个HE有⼀千种指代

⼀千个THE有⼀千种指事

对于注重理解⽂本『意思』的应⽤场景来说

歧义太多

全体stopwords列表   http://www.ranks.nl/stopwords

nltk去除stopwords

⾸先记得在console⾥⾯下载⼀下词库
或者nltk.download(‘stopwords’)

from nltk.corpusimportstopwords
#先token⼀把，得到⼀个word_list
# ...
#然后filter⼀把
filtered_words =
[wordforwordinword_listifwordnot instopwords.words('english')]

四、nltk频率统计　

import nltk
from nltk import FreqDist
# 做个词库先
corpus = 'this is my sentence ' \
           'this is my life ' \
           'this is the day'
# 随便tokenize⼀下
# 显然, 正如上⽂提到,
# 这⾥可以根据需要做任何的preprocessing:
# stopwords, lemma, stemming, etc.
tokens = nltk.word_tokenize(corpus)
print(tokens)
['this', 'is', 'my', 'sentence', 'this', 'is', 'my', 'life', 'this', 'is', 'the', 'day']

　

# 借⽤NLTK的FreqDist统计⼀下⽂字出现的频率
fdist = FreqDist(tokens)
# 它就类似于⼀个Dict
# 带上某个单词, 可以看到它在整个⽂章中出现的次数
print(fdist.most_common(50))
for k,v in fdist.items():
    print(k,v)

[('this', 3), ('is', 3), ('my', 2), ('sentence', 1), ('life', 1), ('the', 1), ('day', 1)]
this 3
is 3
my 2
sentence 1
life 1
the 1
day 1

# 好, 此刻, 我们可以把最常⽤的50个单词拿出来
standard_freq_vector = fdist.most_common(50)
size = len(standard_freq_vector)
print(standard_freq_vector)

[('this', 3), ('is', 3), ('my', 2), ('sentence', 1), ('life', 1), ('the', 1), ('day', 1)]

Func: 按照出现频率⼤⼩, 记录下每⼀个单词的位置　　

def position_lookup(v):
    res = {}
    counter = 0
    for word in v:
        res[word[0]] = counter
        counter += 1
    return res
# 把标准的单词位置记录下来
standard_position_dict = position_lookup(standard_freq_vector)
print(standard_position_dict)
# 得到⼀个位置对照表
{'this': 0, 'is': 1, 'my': 2, 'sentence': 3, 'life': 4, 'the': 5, 'day': 6}

　　这时，我们有个新句子

[1, 1, 0, 0, 0, 0, 0]sentence = 'this is cool'
# 先新建⼀个跟我们的标准vector同样⼤⼩的向量
freq_vector = [0] * size
# 简单的Preprocessing
tokens = nltk.word_tokenize(sentence)
# 对于这个新句⼦⾥的每⼀个单词
for word in tokens:
    try:
        # 如果在我们的词库⾥出现过
        # 那么就在"标准位置"上+1
        freq_vector[standard_position_dict[word]] += 1
    except KeyError:
        # 如果是个新词
        # 就pass掉
        continue
print(freq_vector)
# 第⼀个位置代表 is, 出现了⼀次
# 第⼆个位置代表 this, 出现了⼀次
# 后⾯都⽊有
[1, 1, 0, 0, 0, 0, 0]

　五、nltk实现tf-idf

import nltk
from nltk.text import TextCollection
sents = ['this is sentence one', 'this is sentence two', 'this is sentence three']
sents = [nltk.word_tokenize(sent) for sent in sents]
corpus = TextCollection(sents)

# 直接就能算出tfidf
# (term: ⼀句话中的某个term, text: 这句话)
print(corpus.idf('three'))
print(corpus.tf('four',nltk.word_tokenize('this is a sentence four')))
print(corpus.tf_idf('four',nltk.word_tokenize('this is a sentence four')))

1.0986122886681098
0.2
0.0

　六、svd降维

%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt
la = np.linalg
words = ['I','like','enjoy','deep','learning','NLP','flying']
X = np.array([[0,2,1,0,0,0,0,0],
              [2,0,0,1,0,1,0,0],
              [1,0,0,0,0,0,1,0],
              [0,1,0,0,1,0,0,0],
              [0,0,0,1,0,0,0,1],
              [0,1,0,0,0,0,0,1],
              [0,0,1,0,0,0,0,1],
              [0,0,0,0,1,1,1,0],
             ])
U,s,Vh = la.svd(X,full_matrices=False)
# print(U,s,Vh)
for i in range(len(words)):
    plt.text(U[i,0],U[i,1],words[i])

plt.xlim(-1,1)
plt.ylim(-1,1)
plt.show()

　　

七、nltk经典应用-情感分析

简单的情感分析

import nltk
words = nltk.word_tokenize('I am very happy,i love you')
sentiment_dictionary = {}
for line in open('data/AFINN/AFINN-111.txt'):
    word, score = line.split('\t')
    sentiment_dictionary[word] = int(score)
# 把这个打分表记录在⼀个Dict上以后
# 跑⼀遍整个句⼦，把对应的值相加
total_score = sum(sentiment_dictionary.get(word, 0) for word in words)
# 有值就是Dict中的值，没有就是0
# 于是你就得到了⼀个 sentiment score
print(total_score)
6

　　配上ML的情感分析

from nltk.classify import NaiveBayesClassifier
# 随⼿造点训练集
s1 = 'this is a good book'
s2 = 'this is a awesome book'
s3 = 'this is a bad book'
s4 = 'this is a terrible book'
def preprocess(s):
    # Func: 句⼦处理
    # 这⾥简单的⽤了split(), 把句⼦中每个单词分开
    # 显然 还有更多的processing method可以⽤
    return {word: True for word in s.lower().split()}
    # return⻓这样:
    # {'this': True, 'is':True, 'a':True, 'good':True, 'book':True}
    # 其中, 前⼀个叫fname, 对应每个出现的⽂本单词;
    # 后⼀个叫fval, 指的是每个⽂本单词对应的值。
        # 这⾥我们⽤最简单的True,来表示,这个词『出现在当前的句⼦中』的意义。
    # 当然啦, 我们以后可以升级这个⽅程, 让它带有更加⽜逼的fval, ⽐如 word2vec

# 把训练集给做成标准形式
training_data = [[preprocess(s1), 'pos'],
                 [preprocess(s2), 'pos'],
                 [preprocess(s3), 'neg'],
                 [preprocess(s4), 'neg']]
# 喂给model吃
model = NaiveBayesClassifier.train(training_data)
# 打出结果
print(model.classify(preprocess('this is a good book')))
pos

八、nltk应用-文本相似度