We can apply an off-the-shelf tagger from NLTK or combine multiple taggers to customize the tagging process. It is easy to directly use the built-in tagging function, pos_tag, as in: pos_tag(input_tokens), for instance. But behind the scene, it is actually a prediction from a pre-built supervised learning model. The model is trained based on a large corpus composed of words that are correctly tagged.

Reusing an earlier example, we can perform PoS tagging as follows:

>>> import nltk
>>> tokens = word_tokenize(sent)
>>> print(nltk.pos_tag(tokens))
[('I', 'PRP'), ('am', 'VBP'), ('reading', 'VBG'), ('a', 'DT'), ('book', 'NN'), ('.', '.'), ('It', 'PRP'), ('is', 'VBZ'), ('Python', 'NNP'), ('Machine', 'NNP'), ('Learning', 'NNP'), ('By', 'IN'), ('Example', 'NNP'), (',', ','), ('2nd', 'CD'), ('edition', 'NN'), ('.', '.')]

The PoS tag following each token is returned. We can check the meaning of a tag using the help function. Looking up PRP and VBP, for example, gives us the following output:

>>> nltk.help.upenn_tagset('PRP')
PRP: pronoun, personal
   hers herself him himself hisself it itself me myself one oneself ours ourselves ownself self she thee theirs them themselves they thou thy us
>>> nltk.help.upenn_tagset('VBP')
VBP: verb, present tense, not 3rd person singular
   predominate wrap resort sue twist spill cure lengthen brush terminate appear tend stray glisten obtain comprise detest tease attract emphasize mold postpone sever return wag ...

In spaCy, getting a PoS tag is also easy. The token object parsed from an input sentence has an attribute called pos_, which is the tag we are looking for:

>>> print([(token.text, token.pos_) for token in tokens2])
[('I', 'PRON'), ('have', 'VERB'), ('been', 'VERB'), ('to', 'ADP'), ('U.K.', 'PROPN'), ('and', 'CCONJ'), ('U.S.A.', 'PROPN')]