Installation instructions for NLTK are available at http://nltk.org/install.html and the latest version at the time of writing this is Version 3.0b1. This version of NLTK is built for Python 3.0 or higher, but it is backwards compatible with Python 2.6 and higher. In this book, we will be using Python 3.3.2. If you've used earlier versions of NLTK (such as version 2.0), note that some of the APIs have changed in Version 3 and are not backwards compatible.

Once you've installed NLTK, you'll also need to install the data following the instructions at http://nltk.org/data.html. I recommend installing everything, as we'll be using a number of corpora and pickled objects. The data is installed in a data directory, which on Mac and Linux/Unix is usually /usr/share/nltk_data, or on Windows is C:\nltk_data. Make sure that tokenizers/punkt.zip is in the data directory and has been unpacked so that there's a file at tokenizers/punkt/PY3/english.pickle.

Finally, to run the code examples, you'll need to start a Python console. Instructions on how to do so are available at http://nltk.org/install.html. For Mac and Linux/Unix users, you can open a terminal and type python.

Once NLTK is installed and you have a Python console running, we can start by creating a paragraph of text:

>>> para = "Hello World. It's good to see you. Thanks for buying this book."

Now we want to split the paragraph into sentences. First we need to import the sentence tokenization function, and then we can call it with the paragraph as an argument:

>>> from nltk.tokenize import sent_tokenize
>>> sent_tokenize(para)
['Hello World.', "It's good to see you.", 'Thanks for buying this book.']

So now we have a list of sentences that we can use for further processing.

The sent_tokenize function uses an instance of PunktSentenceTokenizer from the nltk.tokenize.punkt module. This instance has already been trained and works well for many European languages. So it knows what punctuation and characters mark the end of a sentence and the beginning of a new sentence.

The instance used in sent_tokenize() is actually loaded on demand from a pickle file. So if you're going to be tokenizing a lot of sentences, it's more efficient to load the PunktSentenceTokenizer class once, and call its tokenize() method instead:

>>> import nltk.data
>>> tokenizer = nltk.data.load('tokenizers/punkt/PY3/english.pickle')
>>> tokenizer.tokenize(para)
['Hello World.', "It's good to see you.", 'Thanks for buying this book.']

>>> spanish_tokenizer = nltk.data.load('tokenizers/punkt/PY3/spanish.pickle')
>>> spanish_tokenizer.tokenize('Hola amigo. Estoy bien.')
['Hola amigo.', 'Estoy bien.']

In the next recipe, we'll learn how to split sentences into individual words. After that, we'll cover how to use regular expressions to tokenize text. We'll cover how to train your own sentence tokenizer in an upcoming recipe, Training a sentence tokenizer.

Basic word tokenization is very simple; use the word_toke nize() function:

>>> from nltk.tokenize import word_tokenize
>>> word_tokenize('Hello World.')
['Hello', 'World', '.']

The word_tokenize() function is a wrapper function that calls tokenize() on an instance of the TreebankWordTokenizer class. It's equivalent to the following code:

>>> from nltk.tokenize import TreebankWordTokenizer
>>> tokenizer = TreebankWordTokenizer()
>>> tokenizer.tokenize('Hello World.')
['Hello', 'World', '.']

It works by separating words using spaces and punctuation. And as you can see, it does not discard the punctuation, allowing you to decide what to do with it.

Ignoring the obviously named WhitespaceTokenizer and SpaceTokenizer, there are two other word tokenizers worth looking at: PunktWordTokenizer and WordPunctTokenizer. These differ from TreebankWordTokenizer by how they handle punctuation and contractions, but they all inherit from TokenizerI. The inheritance tree looks like what's shown in the following diagram:

>>> word_tokenize("can't")
['ca', "n't"]

>>> from nltk.tokenize import PunktWordTokenizer
>>> tokenizer = PunktWordTokenizer()
>>> tokenizer.tokenize("Can't is a contraction.")
['Can', "'t", 'is', 'a', 'contraction.']

>>> from nltk.tokenize import WordPunctTokenizer
>>> tokenizer = WordPunctTokenizer()
>>> tokenizer.tokenize("Can't is a contraction.")
['Can', "'", 't', 'is', 'a', 'contraction', '.']

For more control over word tokenization, you'll want to read the next recipe to learn how to use regular expressions and the RegexpTokenizer for tokenization. And for more on the Penn Treebank corpus, visit http://www.cis.upenn.edu/~treebank/.

First you need to decide how you want to tokenize a piece of text as this will determine how you construct your regular expression. The choices are:

We'll start with an example of the first, matching alphanumeric tokens plus single quotes so that we don't split up contractions.

We'll create an instance of RegexpTokenizer, giving it a regular expression string to use for matching tokens:

>>> from nltk.tokenize import RegexpTokenizer
>>> tokenizer = RegexpTokenizer("[\w']+")
>>> tokenizer.tokenize("Can't is a contraction.")
["Can't", 'is', 'a', 'contraction']

There's also a simple helper function you can use if you don't want to instantiate the class, as shown in the following code:

>>> from nltk.tokenize import regexp_tokenize
>>> regexp_tokenize("Can't is a contraction.", "[\w']+")
["Can't", 'is', 'a', 'contraction']

Now we finally have something that can treat contractions as whole words, instead of splitting them into tokens.

The RegexpTokenizer class works by compiling your pattern, then calling re.findall() on your text. You could do all this yourself using the re module, but RegexpTokenizer implements the TokenizerI interface, just like all the word tokenizers from the previous recipe. This means it can be used by other parts of the NLTK package, such as corpus readers, which we'll cover in detail in Chapter 3, Creating Custom Corpora. Many corpus readers need a way to tokenize the text they're reading, and can take optional keyword arguments specifying an instance of a TokenizerI subclass. This way, you have the ability to provide your own tokenizer instance if the default tokenizer is unsuitable.

RegexpTokenizer can also work by matching the gaps, as opposed to the tokens. Instead of using re.findall(), the RegexpTokenizer class will use re.split(). This is how the BlanklineTokenizer class in nltk.tokenize is implemented.

>>> tokenizer = RegexpTokenizer('\s+', gaps=True)
>>> tokenizer.tokenize("Can't is a contraction.")
["Can't", 'is', 'a', 'contraction.']

For simpler word tokenization, see the previous recipe.

For this example, we'll be using the webtext corpus, specifically the overheard.txt file, so make sure you've downloaded this corpus. The text in this file is formatted as dialog that looks like this:

White guy: So, do you have any plans for this evening?
Asian girl: Yeah, being angry!
White guy: Oh, that sounds good.

As you can see, this isn't your standard paragraph of sentences formatting, which makes it a perfect case for training a sentence tokenizer.

NLTK provides a PunktSentenceTokenizer class that you can train on raw text to produce a custom sentence tokenizer. You can get raw text either by reading in a file, or from an NLTK corpus using the raw() method. Here's an example of training a sentence tokenizer on dialog text, using overheard.txt from the webtext corpus:

>>> from nltk.tokenize import PunktSentenceTokenizer
>>> from nltk.corpus import webtext
>>> text = webtext.raw('overheard.txt')
>>> sent_tokenizer = PunktSentenceTokenizer(text)

Let's compare the results to the default sentence tokenizer, as follows:

>>> sents1 = sent_tokenizer.tokenize(text)
>>> sents1[0]
'White guy: So, do you have any plans for this evening?'

>>> from nltk.tokenize import sent_tokenize
>>> sents2 = sent_tokenize(text)
>>> sents2[0]
'White guy: So, do you have any plans for this evening?'
>>> sents1[678]
'Girl: But you already have a Big Mac...'
>>> sents2[678]
'Girl: But you already have a Big Mac...\\nHobo: Oh, this is all theatrical.'

While the first sentence is the same, you can see that the tokenizers disagree on how to tokenize sentence 679 (this is the first sentence where the tokenizers diverge). The default tokenizer includes the next line of dialog, while our custom tokenizer correctly thinks that the next line is a separate sentence. This difference is a good demonstration of why it can be useful to train your own sentence tokenizer, especially when your text isn't in the typical paragraph-sentence structure.

The PunktSentenceTokenizer class uses an unsupervised learning algorithm to learn what constitutes a sentence break. It is unsupervised because you don't have to give it any labeled training data, just raw text. You can read more about these kinds of algorithms at https://en.wikipedia.org/wiki/Unsupervised_learning. The specific technique used in this case is called sentence boundary detection and it works by counting punctuation and tokens that commonly end a sentence, such as a period or newline, then using the resulting frequencies to decide what the sentence boundaries should actually look like.

This is a simplified description of the algorithm—if you'd like more details, take a look at the source code of the nltk.tokenize.punkt.PunktTrainer class, which can be found online at http://www.nltk.org/_modules/nltk/tokenize/punkt.html#PunktSentenceTokenizer.

The PunktSentenceTokenizer class learns from any string, which means you can open a text file and read its content. Here is an example of reading overheard.txt directly instead of using the raw() corpus method. This assumes that the webtext corpus is located in the standard directory at /usr/share/nltk_data/corpora. We also have to pass a specific encoding to the open() function, as follows, because the file is not in ASCII:

>>> with open('/usr/share/nltk_data/corpora/webtext/overheard.txt', encoding='ISO-8859-2') as f:
...   text = f.read()
>>> sent_tokenizer = PunktSentenceTokenizer(text)
>>> sents = sent_tokenizer.tokenize(text)
>>> sents[0]
'White guy: So, do you have any plans for this evening?'
>>> sents[678]
'Girl: But you already have a Big Mac...'

Once you have a custom sentence tokenizer, you can use it for your own corpora. Many corpus readers accept a sent_tokenizer parameter, which lets you override the default sentence tokenizer object with your own sentence tokenizer. Corpus readers are covered in more detail in Chapter 3, Creating Custom Corpora.

Most of the time, the default sentence tokenizer will be sufficient. This is covered in the first recipe, Tokenizing text into sentences.

NLTK comes with a stopwords corpus that contains word lists for many languages. Be sure to unzip the data file, so NLTK can find these word lists at nltk_data/corpora/stopwords/.

We're going to create a set of all English stopwords, then use it to filter stopwords from a sentence with the help of the following code:

>>> from nltk.corpus import stopwords
>>> english_stops = set(stopwords.words('english'))
>>> words = ["Can't", 'is', 'a', 'contraction']
>>> [word for word in words if word not in english_stops]
["Can't", 'contraction']

The stopwords corpus is an instance of nltk.corpus.reader.WordListCorpusReader. As such, it has a words() method that can take a single argument for the file ID, which in this case is 'english', referring to a file containing a list of English stopwords. You could also call stopwords.words() with no argument to get a list of all stopwords in every language available.

You can see the list of all English stopwords using stopwords.words('english') or by examining the word list file at nltk_data/corpora/stopwords/english. There are also stopword lists for many other languages. You can see the complete list of languages using the fileids method as follows:

>>> stopwords.fileids()
['danish', 'dutch', 'english', 'finnish', 'french', 'german', 'hungarian', 'italian', 'norwegian', 'portuguese', 'russian', 'spanish', 'swedish', 'turkish']

Any of these fileids can be used as an argument to the words() method to get a list of stopwords for that language. For example:

>>> stopwords.words('dutch')
['de', 'en', 'van', 'ik', 'te', 'dat', 'die', 'in', 'een', 'hij', 'het', 'niet', 'zijn', 'is', 'was', 'op', 'aan', 'met', 'als', 'voor', 'had', 'er', 'maar', 'om', 'hem', 'dan', 'zou', 'of', 'wat', 'mijn', 'men', 'dit', 'zo', 'door', 'over', 'ze', 'zich', 'bij', 'ook', 'tot', 'je', 'mij', 'uit', 'der', 'daar', 'haar', 'naar', 'heb', 'hoe', 'heeft', 'hebben', 'deze', 'u', 'want', 'nog', 'zal', 'me', 'zij', 'nu', 'ge', 'geen', 'omdat', 'iets', 'worden', 'toch', 'al', 'waren', 'veel', 'meer', 'doen', 'toen', 'moet', 'ben', 'zonder', 'kan', 'hun', 'dus', 'alles', 'onder', 'ja', 'eens', 'hier', 'wie', 'werd', 'altijd', 'doch', 'wordt', 'wezen', 'kunnen', 'ons', 'zelf', 'tegen', 'na', 'reeds', 'wil', 'kon', 'niets', 'uw', 'iemand', 'geweest', 'andere']

If you'd like to create your own stopwords corpus, see the Creating a wordlist corpus recipe in Chapter 3, Creating Custom Corpora, to learn how to use WordListCorpusReader. We'll also be using stopwords in the Discovering word collocations recipe later in this chapter.

Be sure you've unzipped the wordnet corpus at nltk_data/corpora/wordnet. This will allow WordNetCorpusReader to access it.

Now we're going to look up the Synset for cookbook, and explore some of the properties and methods of a Synset using the following code:

>>> from nltk.corpus import wordnet
>>> syn = wordnet.synsets('cookbook')[0]
>>> syn.name()
'cookbook.n.01'
>>> syn.definition()
'a book of recipes and cooking directions'

You can look up any word in WordNet using wordnet.synsets(word) to get a list of Synsets. The list may be empty if the word is not found. The list may also have quite a few elements, as some words can have many possible meanings, and, therefore, many Synsets.

Each Synset in the list has a number of methods you can use to learn more about it. The name() method will give you a unique name for the Synset, which you can use to get the Synset directly:

>>> wordnet.synset('cookbook.n.01')
Synset('cookbook.n.01')

The definition() method should be self-explanatory. Some Synsets also have an examples() method, which contains a list of phrases that use the word in context:

>>> wordnet.synsets('cooking')[0].examples()
['cooking can be a great art', 'people are needed who have experience in cookery', 'he left the preparation of meals to his wife']

>>> syn.hypernyms()
[Synset('reference_book.n.01')]
>>> syn.hypernyms()[0].hyponyms()
[Synset('annual.n.02'), Synset('atlas.n.02'), Synset('cookbook.n.01'), Synset('directory.n.01'), Synset('encyclopedia.n.01'), Synset('handbook.n.01'), Synset('instruction_book.n.01'), Synset('source_book.n.01'), Synset('wordbook.n.01')]
>>> syn.root_hypernyms()
[Synset('entity.n.01')]

>>> syn.hypernym_paths()
[[Synset('entity.n.01'), Synset('physical_entity.n.01'), Synset('object.n.01'), Synset('whole.n.02'), Synset('artifact.n.01'), Synset('creation.n.02'), Synset('product.n.02'), Synset('work.n.02'), Synset('publication.n.01'), Synset('book.n.01'), Synset('reference_book.n.01'), Synset('cookbook.n.01')]]

>>> syn.pos()
'n'

>>> len(wordnet.synsets('great'))
7
>>> len(wordnet.synsets('great', pos='n'))
1
>>> len(wordnet.synsets('great', pos='a'))
6

In the next two recipes, we'll explore lemmas and how to calculate Synset similarity. And in Chapter 2, Replacing and Correcting Words, we'll use WordNet for lemmatization, synonym replacement, and then explore the use of antonyms.

In the following code, we'll find that there are two lemmas for the cookbook Synset using the lemmas() method:

>>> from nltk.corpus import wordnet
>>> syn = wordnet.synsets('cookbook')[0]
>>> lemmas = syn.lemmas()
>>> len(lemmas)
2
>>> lemmas[0].name()
'cookbook'
>>> lemmas[1].name()
'cookery_book'
>>> lemmas[0].synset() == lemmas[1].synset()
True

As you can see, cookery_book and cookbook are two distinct lemmas in the same Synset. In fact, a lemma can only belong to a single Synset. In this way, a Synset represents a group of lemmas that all have the same meaning, while a lemma represents a distinct word form.

Since all the lemmas in a Synset have the same meaning, they can be treated as synonyms. So if you wanted to get all synonyms for a Synset, you could do the following:

>>> [lemma.name() for lemma in syn.lemmas()]
['cookbook', 'cookery_book']

>>> synonyms = []
>>> for syn in wordnet.synsets('book'):
...     for lemma in syn.lemmas():
...         synonyms.append(lemma.name())
>>> len(synonyms)
38

>>> len(set(synonyms))
25

>>> gn2 = wordnet.synset('good.n.02')
>>> gn2.definition()
'moral excellence or admirableness'
>>> evil = gn2.lemmas()[0].antonyms()[0]
>>> evil.name
'evil'
>>> evil.synset().definition()
'the quality of being morally wrong in principle or practice'
>>> ga1 = wordnet.synset('good.a.01')
>>> ga1.definition()
'having desirable or positive qualities especially those suitable for a thing specified'
>>> bad = ga1.lemmas()[0].antonyms()[0]
>>> bad.name()
'bad'
>>> bad.synset().definition()
'having undesirable or negative qualities'

In the next recipe, we'll learn how to calculate Synset similarity. Then in Chapter 2, Replacing and Correcting Words, we'll revisit lemmas for lemmatization, synonym replacement, and antonym replacement.

If you were to look at all the hyponyms of reference_book (which is the hypernym of cookbook), you'd see that one of them is instruction_book. This seems intuitively very similar to a cookbook, so let's see what WordNet similarity has to say about it with the help of the following code:

>>> from nltk.corpus import wordnet
>>> cb = wordnet.synset('cookbook.n.01')
>>> ib = wordnet.synset('instruction_book.n.01')
>>> cb.wup_similarity(ib)
0.9166666666666666

So they are over 91% similar!

The wup_similarity method is short for Wu-Palmer Similarity, which is a scoring method based on how similar the word senses are and where the Synsets occur relative to each other in the hypernym tree. One of the core metrics used to calculate similarity is the shortest path distance between the two Synsets and their common hypernym:

>>> ref = cb.hypernyms()[0]
>>> cb.shortest_path_distance(ref)
1
>>> ib.shortest_path_distance(ref)
1
>>> cb.shortest_path_distance(ib)
2

So cookbook and instruction_book must be very similar, because they are only one step away from the same reference_book hypernym, and, therefore, only two steps away from each other.

Let's look at two dissimilar words to see what kind of score we get. We'll compare dog with cookbook, two seemingly very different words.

>>> dog = wordnet.synsets('dog')[0]
>>> dog.wup_similarity(cb)
0.38095238095238093

Wow, dog and cookbook are apparently 38% similar! This is because they share common hypernyms further up the tree:

>>> sorted(dog.common_hypernyms(cb))
[Synset('entity.n.01'), Synset('object.n.01'), Synset('physical_entity.n.01'), Synset('whole.n.02')]

>>> cook = wordnet.synset('cook.v.01')
>>> bake = wordnet.0('bake.v.02')
>>> cook.wup_similarity(bake)
00.6666666666666666

>>> cb.path_similarity(ib)
0.3333333333333333
>>> cb.path_similarity(dog)
0.07142857142857142
>>> cb.lch_similarity(ib)
2.538973871058276
>>> cb.lch_similarity(dog)
0.9985288301111273

The recipe on Looking up Synsets for a word in WordNet has more details about hypernyms and the hypernym tree.

The script for Monty Python and the Holy Grail is found in the webtext corpus, so be sure that it's unzipped at nltk_data/corpora/webtext/.

We're going to create a list of all lowercased words in the text, and then produce BigramCollocationFinder, which we can use to find bigrams, which are pairs of words. These bigrams are found using association measurement functions in the nltk.met rics package, as follows:

>>> from nltk.corpus import webtext
>>> from nltk.collocations import BigramCollocationFinder
>>> from nltk.metrics import BigramAssocMeasures
>>> words = [w.lower() for w in webtext.words('grail.txt')]
>>> bcf = BigramCollocationFinder.from_words(words)
>>> bcf.nbest(BigramAssocMeasures.likelihood_ratio, 4)
[("'", 's'), ('arthur', ':'), ('#', '1'), ("'", 't')]

Well, that's not very useful! Let's refine it a bit by adding a word filter to remove punctuation and stopwords:

>>> from nltk.corpus import stopwords
>>> stopset = set(stopwords.words('english'))
>>> filter_stops = lambda w: len(w) < 3 or w in stopset
>>> bcf.apply_word_filter(filter_stops)
>>> bcf.nbest(BigramAssocMeasures.likelihood_ratio, 4)
[('black', 'knight'), ('clop', 'clop'), ('head', 'knight'), ('mumble', 'mumble')]

Much better, we can clearly see four of the most common bigrams in Monty Python and the Holy Grail. If you'd like to see more than four, simply increase the number to whatever you want, and the collocation finder will do its best.

BigramCollocationFinder constructs two frequency distributions: one for each word, and another for bigrams. A frequency distribution, or FreqDist in NLTK, is basically an enhanced Python dictionary where the keys are what's being counted, and the values are the counts. Any filtering functions that are applied reduce the size of these two FreqDists by eliminating any words that don't pass the filter. By using a filtering function to eliminate all words that are one or two characters, and all English stopwords, we can get a much cleaner result. After filtering, the collocation finder is ready to accept a generic scoring function for finding collocations.

In addition to BigramCollocationFinder, there's also TrigramCollocationFinder, which finds triplets instead of pairs. This time, we'll look for trigrams in Australian singles advertisements with the help of the following code:

>>> from nltk.collocations import TrigramCollocationFinder
>>> from nltk.metrics import TrigramAssocMeasures
>>> words = [w.lower() for w in webtext.words('singles.txt')]
>>> tcf = TrigramCollocationFinder.from_words(words)
>>> tcf.apply_word_filter(filter_stops)
>>> tcf.apply_freq_filter(3)
>>> tcf.nbest(TrigramAssocMeasures.likelihood_ratio, 4)
[('long', 'term', 'relationship')]

Now, we don't know whether people are looking for a long-term relationship or not, but clearly it's an important topic. In addition to the stopword filter, I also applied a frequency filter, which removed any trigrams that occurred less than three times. This is why only one result was returned when we asked for four because there was only one result that occurred more than two times.

The nltk.metrics module will be used again in the Measuring precision and recall of a classifier and Calculating high information words recipes in Chapter 7, Text Classification.