Source Code for Module nltk.book

 1  # Some texts for exploration in chapter 1 of the book 
 2   
 3  from nltk.corpus import gutenberg, genesis, inaugural, nps_chat 
 4  from nltk.text import Text 
 5   
 6  print "Introductory Examples for the NLTK Book" 
 7  print 
 8  print "Loading 5 texts..." 
 9   
10  text1 = Text(gutenberg.words('melville-moby_dick.txt')) 
11  print "text1:", text1.name 
12   
13  text2 = Text(gutenberg.words('austen-sense.txt')) 
14  print "text2:", text2.name 
15   
16  text3 = Text(genesis.words('english-kjv.txt'), name="The Book of Genesis") 
17  print "text3:", text3.name 
18   
19  text4 = Text(inaugural.words(), name="Inaugural Address Corpus") 
20  print "text4:", text4.name 
21   
22  text5 = Text(nps_chat.words(['10-26-teens_706posts.xml', 
23                               '11-08-teens_706posts.xml', 
24                               '11-09-teens_706posts.xml']), 
25                               name="Teen Chat Corpus") 
26  print "text5:", text5.name 
27