文本处理工具 - TextBlob
TextBlob基本介绍
TextBlob是一个用Python编写的开源的文本处理库。它可以用来执行很多自然语言处理的任务,比如,词性标注,名词性成分提取,情感分析,文本翻译,等等。你可以在官方文档阅读TextBlog的所有特性。
基本功能
- Noun phrase extraction 短语提取
- Part-of-speech tagging 词汇标注
- Sentiment analysis 情感分析
- Classification (Naive Bayes, Decision Tree) 分类
- Language translation and detection powered by Google Translate 语言翻译和检查(谷歌翻译支持)
- Tokenization (splitting text into words and sentences) 分词、分句
- Word and phrase frequencies 词、短语频率
- Parsing 语法分析
- n-grams N元标注
- Word inflection (pluralization and singularization) and lemmatization 词反射及词干提取
- Spelling correction 拼写准确性
- Add new models or languages through extensions 添加新模型或语言通过表达
- WordNet integration WordNet整合
快速开始:
Create a TextBlob(创建一个textblob对象)
First, the import. TextBlob 类
>>> from textblob import TextBlob
Let’s create our first TextBlob
.
>>> wiki = TextBlob("Python is a high-level, general-purpose programming language.")
Part-of-speech Tagging(词性标注)
Part-of-speech tags can be accessed through the tags
property.
-
>>> wiki.tags
-
[('Python', 'NNP'), ('is', 'VBZ'), ('a', 'DT'), ('high-level', 'JJ'), ('general-purpose', 'JJ'), ('programming', 'NN'), ('language', 'NN')]
Noun Phrase Extraction(名词短语列表)
Similarly, noun phrases are accessed through the noun_phrases
property. 注意:只提取名词短语
-
>>> wiki.noun_phrases
-
WordList(['python'])
Sentiment Analysis(情感分析)
返回一个元组 Sentiment(polarity, subjectivity)
.
The polarity score is a float within the range [-1.0, 1.0]. -1.0 消极,1.0积极
The subjectivity is a float within the range [0.0, 1.0] 0.0 表示客观,1.0表示主观.
-
>>> testimonial = TextBlob("Textblob is amazingly simple to use. What great fun!")
-
>>> testimonial.sentiment
-
Sentiment(polarity=0.39166666666666666, subjectivity=0.4357142857142857)
-
>>> testimonial.sentiment.polarity
-
0.39166666666666666
Tokenization(分词和分句)
You can break TextBlobs into words or sentences.
-
>>> zen = TextBlob("Beautiful is better than ugly. "
-
... "Explicit is better than implicit. "
-
... "Simple is better than complex.")
-
>>> zen.words
-
WordList(['Beautiful', 'is', 'better', 'than', 'ugly', 'Explicit', 'is', 'better', 'than', 'implicit', 'Simple', 'is', 'better', 'than', 'complex'])
-
>>> zen.sentences
-
[Sentence("Beautiful is better than ugly."), Sentence("Explicit is better than implicit."), Sentence("Simple is better than complex.")]
Sentence
对象 和TextBlobs 一样,有相同的方法和属性.
-
>>> for sentence in zen.sentences:
-
... print(sentence.sentiment)
Words Inflection and Lemmatization(词反射及词干提取:单复数、过去式等)
Each word in TextBlob.words
or Sentence.words
is a Word
object (a subclass of unicode
) with useful methods, e.g. for word inflection.
singularize() 变单数, pluralize()变复数,用在对名词进行处理,且会考虑特殊名词单复数形式
-
>>> sentence = TextBlob('Use 4 spaces per indentation level.')
-
>>> sentence.words
-
WordList(['Use', '4', 'spaces', 'per', 'indentation', 'level'])
-
>>> sentence.words[2].singularize()
-
'space'
-
>>> sentence.words[-1].pluralize()
-
'levels'
Word 类 :lemmatize() 方法 对单词进行词形还原,名词找单数,动词找原型。所以需要一次处理名词,一次处理动词
-
>>> from textblob import Word
-
>>> w = Word("octopi")
-
>>> w.lemmatize() # 默认只处理名词
-
'octopus'
-
>>> w = Word("went")
-
>>> w.lemmatize("v") # 对动词原型处理
-
'go'
WordNet Integration (WordNet整合)
You can access the synsets for a Word
via the synsets
属性 或者用 get_synsets
方法只查看部分或全部synset.
-
>>> from textblob import Word
-
>>> from textblob.wordnet import VERB
-
>>> word = Word("octopus")
-
>>> word.synsets
-
[Synset('octopus.n.01'), Synset('octopus.n.02')]
-
>>> Word("hack").get_synsets(pos=VERB) # 只查找 该词作为 动词 的集合,参数为空时和synsets方法相同
-
[Synset('chop.v.05'), Synset('hack.v.02'), Synset('hack.v.03'), Synset('hack.v.04'), Synset('hack.v.05'), Synset('hack.v.06'), Synset('hack.v.07'), Synset('hack.v.08')]
You can access the definitions for each synset via the definitions
property or the define()
method, which can also take an optional part-of-speech argument.
-
>>> Word("octopus").definitions #单词“章鱼”的定义
-
['tentacles of octopus prepared as food', 'bottom-living cephalopod having a soft oval body with eight long tentacles'] # '章鱼的触手是食物','底硒头足类动物,身体软而呈卵形,有八只长触须'
You can also create synsets directly.
-
>>> from textblob.wordnet import Synset
-
>>> octopus = Synset('octopus.n.02')
-
>>> shrimp = Synset('shrimp.n.03')
-
>>> octopus.path_similarity(shrimp)
-
0.1111111111111111
For more information on the WordNet API, see the NLTK documentation on the Wordnet Interface.
WordLists
A WordList
is just a Python list with additional methods. 属性words : 一个包含句子分词的list
-
>>> animals = TextBlob("cat dog octopus")
-
>>> animals.words
-
WordList(['cat', 'dog', 'octopus'])
-
>>> animals.words.pluralize()
-
WordList(['cats', 'dogs', 'octopodes'])
Spelling Correction(拼写校正)
Use the correct()
method to attempt spelling correction.
-
>>> b = TextBlob("I havv goood speling!")
-
>>> print(b.correct())
-
I have good spelling!
Word
objects have a spellcheck() Word.spellcheck()
method that returns a list of (word,confidence)
tuples with spelling suggestions.
-
>>> from textblob import Word
-
>>> w = Word('falibility')
-
>>> w.spellcheck()
-
[('fallibility', 1.0)]
Spelling correction is based on Peter Norvig’s “How to Write a Spelling Corrector”[1] as implemented in the pattern library. It is about 70% accurate [2].
Get Word and Noun Phrase Frequencies(单词词频)
There are two ways to get the frequency of a word or noun phrase in a TextBlob
. 两种方法来获取单词频次
The first is through the word_counts
dictionary. 从属性word_counts 字典获取
-
>>> monty = TextBlob("We are no longer the Knights who say Ni. "
-
... "We are now the Knights who say Ekki ekki ekki PTANG.")
-
>>> monty.word_counts['ekki']
-
3
If you access the frequencies this way, the search will not be case sensitive, and words that are not found will have a frequency of 0.
The second way is to use the count()
method. 用count ()方法获取
-
>>> monty.words.count('ekki') #单词频次
-
3
You can specify whether or not the search should be case-sensitive (default is False
).
-
>>> monty.words.count('ekki', case_sensitive=True) #设置大小写敏感,默认不区分
-
2
Each of these methods can also be used with noun phrases.
-
>>> wiki.noun_phrases.count('python') #短语频次
-
1
Translation and Language Detection(翻译及语言检测语言)
New in version 0.5.0
.
TextBlobs can be translated between languages.
-
>>> en_blob = TextBlob(u'Simple is better than complex.')
-
>>> en_blob.translate(to='es')
-
TextBlob("Simple es mejor que complejo.")
If no source language is specified, TextBlob will attempt to detect the language. You can specify the source language explicitly, like so. Raises TranslatorError
if the TextBlob cannot be translated into the requested language or NotTranslated
if the translated result is the same as the input string.
-
>>> chinese_blob = TextBlob(u"美丽优于丑陋")
-
>>> chinese_blob.translate(from_lang="zh-CN", to='en')
-
TextBlob("Beautiful is better than ugly")
You can also attempt to detect a TextBlob’s language using TextBlob.detect_language()
.
-
>>> b = TextBlob(u"بسيط هو أفضل من مجمع")
-
>>> b.detect_language()
-
'ar'
As a reference, language codes can be found here.
Language translation and detection is powered by the Google Translate API.
Parsing(解析)
Use the parse()
method to parse the text. 句法解析 parse() 方法
-
>>> b = TextBlob("And now for something completely different.")
-
>>> print(b.parse())
-
And/CC/O/O now/RB/B-ADVP/O for/IN/B-PP/B-PNP something/NN/B-NP/I-PNP completely/RB/B-ADJP/O different/JJ/I-ADJP/O ././O/O
By default, TextBlob uses pattern’s parser [3].
TextBlobs Are Like Python Strings!(TextBlobs像是字符串)
You can use Python’s substring syntax.
-
>>> zen[0:19]
-
TextBlob("Beautiful is better")
You can use common string methods.
-
>>> zen.upper()
-
TextBlob("BEAUTIFUL IS BETTER THAN UGLY. EXPLICIT IS BETTER THAN IMPLICIT. SIMPLE IS BETTER THAN COMPLEX.")
-
>>> zen.find("Simple")
-
65
You can make comparisons between TextBlobs and strings.
-
>>> apple_blob = TextBlob('apples')
-
>>> banana_blob = TextBlob('bananas')
-
>>> apple_blob < banana_blob
-
True
-
>>> apple_blob == 'apples'
-
True
You can concatenate and interpolate TextBlobs and strings.
-
>>> apple_blob + ' and ' + banana_blob
-
TextBlob("apples and bananas")
-
>>> "{0} and {1}".format(apple_blob, banana_blob)
-
'apples and bananas'
n
-grams(提取前n个字)
The TextBlob.ngrams()
method returns a list of tuples of n
successive words.
ngrams(n) 方法返回 句子每 n 个连续单词为一个元素的 list
-
>>> blob = TextBlob("Now is better than never.")
-
>>> blob.ngrams(n=3)
-
[WordList(['Now', 'is', 'better']), WordList(['is', 'better', 'than']), WordList(['better', 'than', 'never'])]
Get Start and End Indices of Sentences(句子开始和结束的索引)
Use sentence.start
and sentence.end
to get the indices where a sentence starts and ends within a TextBlob
.
-
>>> for s in zen.sentences:
-
... print(s)
-
... print("---- Starts at index {}, Ends at index {}".format(s.start, s.end))
-
Beautiful is better than ugly.
-
---- Starts at index 0, Ends at index 30
-
Explicit is better than implicit.
-
---- Starts at index 31, Ends at index 64
-
Simple is better than complex.
-
---- Starts at index 65, Ends at index 95
文档
TextBlob is a Python library for processing textual data. It provides a simple API for diving into common (NLP) tasks such as part-of-speech tagging, noun phrase extraction, sentiment analysis, classification, translation, and more.
[html] view plain copy
- from textblob import TextBlob
- text = '''
- The titular threat of The Blob has always struck me as the ultimate movie
- monster: an insatiably hungry, amoeba-like mass able to penetrate
- virtually any safeguard, capable of--as a doomed doctor chillingly
- describes it--"assimilating flesh on contact.
- Snide comparisons to gelatin be damned, it's a concept with the most
- devastating of potential consequences, not unlike the grey goo scenario
- proposed by technological theorists fearful of
- artificial intelligence run rampant.
- '''
- blob = TextBlob(text)
- blob.tags # [('The', 'DT'), ('titular', 'JJ'),
- # ('threat', 'NN'), ('of', 'IN'), ...]
- blob.noun_phrases # WordList(['titular threat', 'blob',
- # 'ultimate movie monster',
- # 'amoeba-like mass', ...])
- for sentence in blob.sentences:
- print(sentence.sentiment.polarity)
- # 0.060
- # -0.341
- blob.translate(to="es") # 'La amenaza titular de The Blob...
TextBlob stands on the giant shoulders of NLTK and pattern, and plays nicely with both.
Features
- Noun phrase extraction
- Part-of-speech tagging
- Sentiment analysis
- Classification (Naive Bayes, Decision Tree)
- Language translation and detection powered by Google Translate
- Tokenization (splitting text into words and sentences)
- Word and phrase frequencies
- Parsing
- n-grams
- Word inflection (pluralization and singularization) and lemmatization
- Spelling correction
- Add new models or languages through extensions
- WordNet integration