Python – Chunking and Classification

Classification-based chunking involves classifying text into groups of words rather than individual words. A simple scenario is tokenizing text into sentences. We will use a corpus to demonstrate classification. We have chosen the conll2000 corpus, which contains data from the Wall Street Journal Corpus (WSJ) for noun phrase chunking.

First, we add the corpus to our environment using the following command.

import nltk
nltk.download('conll2000')

Let’s take a look at the first few sentences in this corpus.

from nltk.corpus import conll2000

x = (conll2000.sents())
for i in range(3):
print x[i]
print 'n'

When we run the above program, we get the following output −

['Confidence', 'in', 'the', 'pond', 'is', 'widely', 'expected', 'to', 'take', 'another', 'sharp', 'dive', 'if', 'trade', 'figres', 'for', 'September', ',', 'de', 'for', 'release', 'tomorrow', ',', 'fail', 'to', 'show', 'a', 'sbstantial', 'improvement', 'from', 'Jly', 'and', 'Agst', "'s", 'near-record', 'deficits', '.']


['Chancellor', 'of', 'the', 'Excheqer', 'Nigel', 'Lawson', "'s", 'restated', 'commitment', 'to', 'a', 'firm', 'monetary', 'policy', 'has', 'helped', 'to', 'prevent', 'a', 'freefall', 'in', 'sterling', 'over', 'the', 'past', 'week', '.']

['Bt', 'analysts', 'reckon', 'nderlying', 'spport', 'for', 'sterling', 'has', 'been', 'eroded', 'by', 'the', 'chancellor', "'s", 'failre', 'to', 'annonce', 'any', 'new', 'policy', 'measres', 'in', 'his', 'Mansion', 'Hose', 'speech', 'last', 'Thrsday', '.']

Next, we use the tagged_sents() function to tag the sentences with their classifiers.

from nltk.corpus import conll2000

x = (conll2000.tagged_sents())
for i in range(3):
print x[i]
print 'n'

When we run the above program, we get the following output −

[('Confidence', 'NN'), ('in', 'IN'), ('the', 'DT'), ('pond', 'NN'), ('is', 'VBZ'), ('widely', 'RB'), ('expected', 'VBN'), ('to', 'TO'), ('take', 'VB'), ('another', 'DT'), ('sharp', 'JJ'), ('dive', 'NN'), ('if', 'IN'), ('trade', 'NN'), ('figres', 'NNS'), ('for', 'IN'), ('September', 'NNP'), (',', ','), ('de', 'JJ'), ('for', 'IN'), ('release', 'NN'), ('tomorrow', 'NN'), (',', ','), ('fail', 'VB'), ('to', 'TO'), ('show', 'VB'), ('a', 'DT'), ('sbstantial', 'JJ'), ('improvement', 'NN'), ('from', 'IN'), ('Jly', 'NNP'), ('and', 'CC'), ('Agst', 'NNP'), ("'s", 'POS'), ('near-record', 'JJ'), ('deficits', 'NNS'), ('.', '.')]


[('Chancellor', 'NNP'), ('of', 'IN'), ('the', 'DT'), ('Excheqer', 'NNP'), ('Nigel', 'NNP'), ('Lawson', 'NNP'), ("'s", 'POS'), ('restated', 'VBN'), ('commitment', 'NN'), ('to', 'TO'), ('a', 'DT'), ('firm', 'NN'), ('monetary', 'JJ'), ('policy', 'NN'), ('has', 'VBZ'), ('helped', 'VBN'), ('to', 'TO'), ('prevent', 'VB'), ('a', 'DT'), ('freefall', 'NN'), ('in', 'IN'), ('sterling', 'NN'), ('over', 'IN'), ('the', 'DT'), ('past', 'JJ'), ('week', 'NN'), ('.', '.')]


[('Bt', 'CC'), ('analysts', 'NNS'), ('reckon', 'VBP'), ('nderlying', 'VBG'), ('spport', 'NN'), ('for', 'IN'), ('sterling', 'NN'), ('has', 'VBZ'), ('been', 'VBN'), ('eroded', 'VBN'), ('by', 'IN'), ('the', 'DT'), ('chancellor', 'NN'), ("'s", 'POS'), ('failre', 'NN'), ('to', 'TO'), ('annonce', 'VB'), ('any', 'DT'), ('new', 'JJ'), ('policy', 'NN'), ('measres', 'NNS'), ('in', 'IN'), ('his', 'PRP$'), ('Mansion', 'NNP'), ('Hose', 'NNP'), ('speech', 'NN'), ('last', 'JJ'), ('Thrsday', 'NNP'), ('.', '.')]