Source Code for Module nltk.draw.dispersion

 1  # Natural Language Toolkit: Dispersion Plots 
 2  # 
 3  # Copyright (C) 2001-2011 NLTK Project 
 4  # Author: Steven Bird <sb@csse.unimelb.edu.au> 
 5  # URL: <http://www.nltk.org/> 
 6  # For license information, see LICENSE.TXT 
 7   
 8  """ 
 9  A utility for displaying lexical dispersion. 
10  """ 
11   
12 -def dispersion_plot(text, words): 
13      """ 
14      Generate a lexical dispersion plot. 
15   
16      @param text: The source text 
17      @type text: C{list} or C{enum} of C{str} 
18      @param words: The target words 
19      @type words: C{list} of C{str} 
20      """ 
21   
22      try: 
23          import pylab 
24      except ImportError: 
25          raise ValueError('The plot function requires the matplotlib package (aka pylab).' 
26                       'See http://matplotlib.sourceforge.net/') 
27   
28      text = list(text) 
29      words.reverse() 
30      points = [(x,y) for x in range(len(text)) 
31                      for y in range(len(words)) 
32                      if text[x] == words[y]] 
33      if points: 
34          x, y = zip(*points) 
35      else: 
36          x = y = () 
37      pylab.plot(x, y, "b|", scalex=.1) 
38      pylab.yticks(range(len(words)), words, color="b") 
39      pylab.ylim(-1, len(words)) 
40      pylab.title("Lexical Dispersion Plot") 
41      pylab.xlabel("Word Offset") 
42      pylab.show() 
43   
44  if __name__ == '__main__': 
45      from nltk.corpus import gutenberg 
46      words = ['Elinor', 'Marianne', 'Edward', 'Willoughby'] 
47      dispersion_plot(gutenberg.words('austen-sense.txt'), words) 
48