Exploring graph properties of the Twitter stream with twython, networkx and IPython

text_utils.py

"""
Utilities for simple text analysis: word frequencies and co-occurrence graph.

These tools can be used to analyze a plain text file treating it as a list of
newline-separated sentences (e.g. a list of paper titles).

It computes word frequencies (after doing some naive normalization by
lowercasing and throwing away a few overly common words).  It also computes,
from the most common words, a weighted graph of word co-occurrences and
displays it, as well as summarizing the graph structure by ranking its nodes in
descending order of eigenvector centrality.

This is meant as an illustration of text processing in Python, using matplotlib
for visualization and NetworkX for graph-theoretical manipulation.  It should
not be considered production-strength code for serious text analysis.

Author: Fernando Perez <[email protected]>
"""

#-----------------------------------------------------------------------------
# Imports
#-----------------------------------------------------------------------------

# From the standard library
import os
import re
import urllib2

# Third-party libraries
import networkx as nx
import numpy as np

from matplotlib import pyplot as plt

#-----------------------------------------------------------------------------
# Function definitions
#-----------------------------------------------------------------------------

def rescale_arr(arr, amin, amax):
    """Rescale an array to a new range.

    Return a new array whose range of values is (amin, amax).

    Parameters
    ----------
    arr : array-like

    amin : float
      new minimum value

    amax : float
      new maximum value

    Examples
    --------
    >>> a = np.arange(5)

    >>> rescale_arr(a,3,6)
    array([ 3.  ,  3.75,  4.5 ,  5.25,  6.  ])
    """

    # old bounds
    m = arr.min()
    M = arr.max()
    # scale/offset
    s = float(amax-amin)/(M-m)
    d = amin - s*m

    # Apply clip before returning to cut off possible overflows outside the
    # intended range due to roundoff error, so that we can absolutely guarantee
    # that on output, there are no values > amax or < amin.
    return np.clip(s*arr+d,amin,amax)


def all_pairs(items):
    """Make all unique pairs (order doesn't matter)"""
    pairs = []
    nitems = len(items)
    for i, wi in enumerate(items):
        for j in range(i+1, nitems):
            pairs.append((wi, items[j]))
    return pairs


def removal_set(words, query):
    """Create a set of words for removal for a given query."""
    rem = set(words.split())
    qw = [w.lower()  for w in query.split()]
    for w in qw:
        rem.add(w)
        rem.add('#' + w)
        qq = ''.join(qw)
        rem.add(qq)
        rem.add('#' + qq)
    return rem

def lines_cleanup(lines, min_length=4, remove = None):
    """Clean up a list of lowercase strings of text for simple analysis.

    Splits on whitespace, removes all 'words' less than `min_length` characters
    long, and those in the `remove` set.

    Returns a list of strings.
    """
    remove = set(remove) if remove is not None else []
    filtered = []
    for line in lines:
        a = []
        for w in line.lower().split():
            wnorm = w.rstrip('.,:').replace('[', '').replace(']', '')
            if len(wnorm) >= min_length and wnorm not in remove:
                a.append(wnorm)
        filtered.append(' '.join(a))
    return filtered


def print_vk(lst):
    """Print a list of value/key pairs nicely formatted in key/value order."""

    # Find the longest key: remember, the list has value/key paris, so the key
    # is element [1], not [0]
    longest_key = max([len(word) for word, count in lst])
    # Make a format string out of it
    fmt = '%'+str(longest_key)+'s -> %s'
    # Do actual printing
    for k,v in lst:
        print fmt % (k,v)


def word_freq(text):
    """Return a dictionary of word frequencies for the given text.

    Input text should be given as an iterable of strings."""

    freqs = {}
    for word in text:
        freqs[word] = freqs.get(word, 0) + 1
    return freqs


def sort_freqs(freqs):
    """Sort a word frequency histogram represented as a dictionary.

    Parameters
    ----------
    freqs : dict
      A dict with string keys and integer values.

    Return
    ------
    items : list
      A list of (count, word) pairs.
    """
    items = freqs.items()
    items.sort(key = lambda wc: wc[1])
    return items


def summarize_freq_hist(freqs, n=10):
    """Print a simple summary of a word frequencies dictionary.

    Paramters
    ---------
    freqs : dict or list
      Word frequencies, represented either as a dict of word->count, or as a
      list of count->word pairs.

    n : int
      The number of least/most frequent words to print.
    """

    items = sort_freqs(freqs) if isinstance(freqs, dict) else freqs
    print 'Number of unique words:',len(freqs)
    print
    print '%d least frequent words:' % n
    print_vk(items[:n])
    print
    print '%d most frequent words:' % n
    print_vk(items[-n:])


def co_occurrences(lines, words):
    """Return histogram of co-occurrences of words in a list of lines.

    Parameters
    ----------
    lines : list
      A list of strings considered as 'sentences' to search for co-occurrences.

    words : list
      A list of words from which all unordered pairs will be constructed and
      searched for co-occurrences.
    """
    wpairs = all_pairs(words)

    # Now build histogram of co-occurrences
    co_occur = {}
    for w1, w2 in wpairs:
        rx = re.compile('%s .*%s|%s .*%s' % (w1, w2, w2, w1))
        co_occur[w1, w2] = sum([1 for line in lines if rx.search(line)])

    return co_occur


def co_occurrences_graph(word_hist, co_occur, cutoff=0):
    """Convert a word histogram with co-occurrences to a weighted graph.

    Edges are only added if the count is above cutoff.
    """
    g = nx.Graph()
    for word, count in word_hist:
        g.add_node(word, count=count)
    for (w1, w2), count in co_occur.iteritems():
        if count<=cutoff:
            continue
        g.add_edge(w1, w2, weight=count)
    return g

# Hack: offset the most central node to avoid too much overlap
rad0 = 0.3

def centrality_layout(wgraph, centrality):
    """Compute a layout based on centrality.
    """
    # Create a list of centralities, sorted by centrality value
    cent = sorted(centrality.items(), key=lambda x:float(x[1]), reverse=True)
    nodes = [c[0] for c in cent]
    cent  = np.array([float(c[1]) for c in cent])
    rad = (cent - cent[0])/(cent[-1]-cent[0])
    rad = rescale_arr(rad, rad0, 1)
    angles = np.linspace(0, 2*np.pi, len(centrality))
    layout = {}
    for n, node in enumerate(nodes):
        r = rad[n]
        th = angles[n]
        layout[node] = r*np.cos(th), r*np.sin(th)
    return layout


def plot_graph(wgraph, pos=None, fig=None, title=None):
    """Conveniently summarize graph visually"""

    # config parameters
    edge_min_width= 3
    edge_max_width= 12
    label_font = 18
    node_font = 22
    node_alpha = 0.4
    edge_alpha = 0.55
    edge_cmap = plt.cm.Spectral

    # Create figure
    if fig is None:
        fig, ax = plt.subplots()
    else:
        ax = fig.add_subplot(111)
    fig.subplots_adjust(0,0,1)

    # Plot nodes with size according to count
    sizes = []
    degrees = []
    for n, d in wgraph.nodes_iter(data=True):
        sizes.append(d['count'])
        degrees.append(wgraph.degree(n))

    sizes = rescale_arr(np.array(sizes, dtype=float), 100, 1000)

    # Compute layout and label edges according to weight
    pos = nx.spring_layout(wgraph) if pos is None else pos
    labels = {}
    width = []
    for n1, n2, d in wgraph.edges_iter(data=True):
        w = d['weight']
        labels[n1, n2] = w
        width.append(w)

    width = rescale_arr(np.array(width, dtype=float), edge_min_width, 
                        edge_max_width)

    # Draw
    nx.draw_networkx_nodes(wgraph, pos, node_size=sizes, node_color=degrees,
                           alpha=node_alpha)
    nx.draw_networkx_edges(wgraph, pos, width=width, edge_color=width,
                           edge_cmap=edge_cmap, alpha=edge_alpha)
    nx.draw_networkx_edge_labels(wgraph, pos, edge_labels=labels, 
                                 font_size=label_font)
    nx.draw_networkx_labels(wgraph, pos, font_size=node_font, font_weight='bold')
    if title is not None:
        ax.set_title(title, fontsize=label_font)
    ax.set_xticks([])
    ax.set_yticks([])

    # Mark centrality axes
    kw = dict(color='k', linestyle='-')
    cross = [ax.axhline(0, **kw), ax.axvline(rad0, **kw)]
    [ l.set_zorder(0) for l in cross]


def plot_word_histogram(freqs, show=10, title=None):
    """Plot a histogram of word frequencies, limited to the top `show` ones.
    """
    sorted_f = sort_freqs(freqs) if isinstance(freqs, dict) else freqs

    # Don't show the tail
    if isinstance(show, int):
        # interpret as number of words to show in histogram
        show_f = sorted_f[-show:]
    else:
        # interpret as a fraction
        start = -int(round(show*len(freqs)))
        show_f = sorted_f[start:]

    # Now, extract words and counts, plot
    n_words = len(show_f)
    ind = np.arange(n_words)
    words = [i[0] for i in show_f]
    counts = [i[1] for i in show_f]

    fig = plt.figure()
    ax = fig.add_subplot(111)

    if n_words<=20:
        # Only show bars and x labels for small histograms, they don't make
        # sense otherwise
        ax.bar(ind, counts)
        ax.set_xticks(ind)
        ax.set_xticklabels(words, rotation=45)
        fig.subplots_adjust(bottom=0.25)
    else:
        # For larger ones, do a step plot
        ax.step(ind, counts)

    # If it spans more than two decades, use a log scale
    if float(max(counts))/min(counts) > 100:
        ax.set_yscale('log')

    if title:
        ax.set_title(title)
    return ax


def summarize_centrality(centrality):
    c = centrality.items()
    c.sort(key=lambda x:x[1], reverse=True)
    print '\nGraph centrality'
    for node, cent in c:
        print "%15s: %.3g" % (node, float(cent))

TwitterGraphs.ipynb

{
 "metadata": {
  "name": "TwitterGraphs"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "Exploring graph properties of the Twitter stream"
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "An interactive narrative built with twython, NetworkX and IPython"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Author:** [Fernando Perez](http://fperez.org), [@fperez_org](http://twitter.com/fperez_org).\n",
      "\n",
      "**Location:** this example is available as a [public Gist](https://gist.github.com/fperez/5681541) that contains both this notebook and the accompanying Python `text_utils.py` module.\n",
      "\n",
      "In this example, we use the IPython notebook to mine data from Twitter with the [twython library](http://pypi.python.org/pypi/twython).  Once we have fetched the raw stream for a specific query, we will at first do some basic word frequency analysis on the results using Python's builtin dictionaries, and then we will use the excellent [NetworkX](http://networkx.lanl.gov) library developed at Los Alamos National Laboratory to look at the results as a network and understand some of its properties.  \n",
      "\n",
      "Using NetworkX, we aim to answer the following questions: for a given query, which words tend to appear together in tweets, and global pattern of relationships between these words emerges from the entire set of results?\n",
      "\n",
      "Obviously the analysis of text corpora of this kind is a complex topic at the intersection of natural language processing, graph theory and statistics, and here we do not pretend to provide an exhaustive coverage of it.  Rather, we want to show you how with a small amount of easy to write code, it is possible to do a few non-trivial things based on real-time data from the Twitter stream.  Hopefully this will serve as a good starting point;  for further reading you can find in-depth discussions of analysing social network data in Python in the book [Mining the Social Web](http://shop.oreilly.com/product/0636920010203.do).\n",
      "\n",
      "**Note:** for this you'll need to have NetworkX as well as twython installed.  You can do so in Ubuntu with the command\n",
      "\n",
      "    sudo apt-get install python-networkx\n",
      "    sudo pip install twython\n",
      "\n",
      "or in other systems with\n",
      "\n",
      "    sudo pip install networkx\n",
      "    sudo pip install twython\n",
      "\n",
      "where you should simply omit the `sudo` command if you are using Windows."
     ]
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Initialization and libraries"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We start by loading the pylab plot support as well as importing NetworkX and the free Twython library to query Twitter's stream:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%pylab inline\n",
      "import networkx as nx\n",
      "from twython import Twython"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Welcome to pylab, a matplotlib-based Python environment [backend: module://IPython.kernel.zmq.pylab.backend_inline].\n",
        "For more information, type 'help(pylab)'.\n"
       ]
      }
     ],
     "prompt_number": 22
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now, we load a local library with some analysis utilities whose code is a bit long to display inline. If you downloaded the [complete gist](https://gist.github.com/fperez/5681541) you should already have it."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "import text_utils as tu  # shorthand for convenience"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "And we create the main Twitter object we'll use later for all queries:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "twitter = Twython()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Query declaration"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here we define which query we want to perform, as well as which words we want to filter out from our analysis because they appear very commonly and we're not interested in them.  \n",
      "\n",
      "Typically you want to run the query once, and after seeing what comes out, fine-tune the removal list, as the definition of which words are considered 'noise' is fairly query-specific (and also changes over time, depending on what's happening out there on Twitter):"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "query = \"Big Data\"\n",
      "words_to_remove = \"\"\"with what some your just have from it's /via &\n",
      "that they your there this\"\"\""
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 12
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Perform query to Twitter servers"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This is the cell that actually fetches data from Twitter.  We limit the output to the first 30 pages of search max (though typically Twitter stops returning results before that)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "n_pages = 30\n",
      "\n",
      "results = []\n",
      "retweets = []\n",
      "for page in range(1, n_pages+1):\n",
      "    search = twitter.search(q=query+' lang:en', page=str(page))\n",
      "    res = search['results']\n",
      "    if not res:\n",
      "        print 'No more results returned, stopping at page:', page\n",
      "        break\n",
      "        \n",
      "    for t in res:\n",
      "        if t['text'].startswith('RT '):\n",
      "            retweets.append(t)\n",
      "        else:\n",
      "            results.append(t)\n",
      "            \n",
      "tweets = [t['text'] for t in results]\n",
      "\n",
      "# Quick summary\n",
      "print 'Query:   ', query\n",
      "print 'Results: ', len(results)\n",
      "print 'Retweets:', len(retweets)\n",
      "print 'Variable `tweets` has a list of all the tweet texts'"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "No more results returned, stopping at page: 23\n",
        "Query:    Big Data\n",
        "Results:  219\n",
        "Retweets: 111\n",
        "Variable `tweets` has a list of all the tweet texts\n"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Text statistics"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's see what the first 10 tweets look like:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tweets[:10]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 13,
       "text": [
        "[u'If enough people played the game, they could really muck up BIG DATA, with wrong information. Yes!',\n",
        " u'Most girls with big bums are facially challenged, there is statistical data to prove this.',\n",
        " u'The Real Reason #Hadoop Is Such a Big Deal in #BigData http://t.co/H3kAyFl3mV via @TheTechScribe, @RWW',\n",
        " u'@Mark_Goldberg Hey Mark, what are your top 3 concerns for Big Data? Take the poll.\\nhttp://t.co/XtUmoPXVev',\n",
        " u\"Weird! RT A Big Day Out at... the Guardian's New 'Data-Driven' Coffee Shop! | VICE United Kingdom http://t.co/C3VyMNuSXH via @VICEUK\",\n",
        " u'CW500: Forget the fancy graphics: how to make big data work for ordinary people \\xa0 http://t.co/ge5Xhi3KDN #b2b',\n",
        " u'Big data needs thick data - http://t.co/7q3HZRsOkC',\n",
        " u'Big Data And Simulations Are Transforming Marketing http://t.co/f8LpZrlDlm\"',\n",
        " u'Without Analytics, Big Data is Just Noise @briansolis - http://t.co/wW1WFKOnKF',\n",
        " u'Toyota announces upcoming launch in Japan of the Big Data Traffic Information Service http://t.co/N9j0xSojnk']"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now we do some cleanup of the common words above, so that we can then compute some basic statistics:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "remove = tu.removal_set(words_to_remove, query)\n",
      "lines = tu.lines_cleanup([tweet['text'].encode('utf-8') for tweet in results], remove=remove)\n",
      "words = '\\n'.join(lines).split()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 14
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Compute frequency histogram:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "wf = tu.word_freq(words)\n",
      "sorted_wf = tu.sort_freqs(wf)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 15
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's look at a summary of the word frequencies from this dataset:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tu.summarize_freq_hist(sorted_wf)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Number of unique words: 1138\n",
        "\n",
        "10 least frequent words:\n",
        " http://t.co/o68p8nbtnl -> 1\n",
        "                   shop -> 1\n",
        "                managed -> 1\n",
        " http://t.co/obrc49sxle -> 1\n",
        "              @rnrworks -> 1\n",
        "https://t.co/5wosulam2t -> 1\n",
        "                 #aapor -> 1\n",
        "                   four -> 1\n",
        "              avalanche -> 1\n",
        "                 @klout -> 1\n",
        "\n",
        "10 most frequent words:\n",
        "       real -> 13\n",
        "       like -> 13\n",
        "     tracks -> 14\n",
        "   ventures -> 14\n",
        "  discovery -> 14\n",
        "    obama's -> 14\n",
        "      funds -> 14\n",
        "#humanswarm -> 16\n",
        "  analytics -> 17\n",
        "     google -> 17\n"
       ]
      }
     ],
     "prompt_number": 16
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now we can plot the histogram of the `n_words` most frequent words:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "n_words = 15\n",
      "tu.plot_word_histogram(sorted_wf, n_words,\"Frequencies for %s most frequent words\" % n_words);"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAEXCAYAAACNj66GAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x496e490>"
       ]
      }
     ],
     "prompt_number": 17
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Above we trimmed the historgram to only show `n_words` because the distribution is very sharply peaked; this is what the histogram for the whole word list looks like:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tu.plot_word_histogram(sorted_wf, 1.0, \"Frequencies for entire word list\");"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XtYVHX+B/D3ICReBrkIgzogyEVAEFE0I5GxRDc31JR6\nxNICV7eLbZaZj+uaPPh476buVlupoG66rbWruUpmOOjmlVCzbNOUQRQ1byiIXITP7w9+zTqCXGYO\nDpzer+fxYeZcvufzGfDN4cw5ZzQiIiAiItVxsHcBRETUPBjwREQqxYAnIlIpBjwRkUox4ImIVIoB\nT0SkUgx4ajanT5+GVquF0mfi/vOf/4SPjw+0Wi2OHDmi6NhK0Gq1MJlM9i6jSVJTUzFhwoQ65xmN\nRvj4+Jifh4eHY9euXfeqNLIBA74F8PPzQ/v27aHVaqHVauHi4oLz58/buyyb+fr6ori4GBqNRtFx\nX331Vbz77rsoLi5GZGSkomM3lcFgwMqVKy2mFRcXw8/Pzz4FWakp36PvvvsOgwcPrncZk8kEBwcH\nVFdX21oa2YAB3wJoNBps2bIFxcXFKC4uxvXr1+Ht7W2xzK1bt+xUXcsiIjh9+jTCwsKsWl/pwGnq\nL6+WEHhVVVX3bFu8jtK+GPAtmIODA959910EBQWhZ8+eAIAtW7agT58+cHNzw4MPPoijR4+alz90\n6BD69u0LFxcXjBs3DuPGjcOcOXMAAOnp6YiNja01/qlTpwAA5eXlePXVV9G9e3d4e3vjueeeQ1lZ\nGYCaP9H1ej3eeust6HQ6dO3aFenp6eZxbt68ienTp8PPzw+urq6IjY1FeXl5rb24a9euYdKkSeja\ntSv0ej3mzJljnvfTTz8hLi4Orq6u8PT0xLhx42q9HuXl5dBqtaiqqkJkZCSCgoIAAD/88AMMBgPc\n3NwQHh6Ozz//3LzOM888g+eeew4jRoxAx44dYTQaa41bX13p6ekYNGgQZsyYAXd3d/To0QOZmZkA\ngNmzZ2P37t2YOnUqtFot/vCHP9R6XevafmFhIcaOHQsvLy/06NEDK1asqPP7n5eXBzc3N/PzyZMn\nQ6fTmZ9PmDABy5YtAwAUFhZi5MiR8PDwQFBQED766CPzcqmpqUhMTMSECRPQqVMnZGRkIC8vD3Fx\ncXBxccGwYcNw6dKlOmuoi5+fH7KysgAABw4cQHR0NDp16gRvb2+8+uqrAGDew3d1dYVWq8X+/fsb\nPT4pSMju/Pz8ZMeOHbWmazQaGTZsmFy9elXKysokNzdXvLy85MCBA1JdXS0ZGRni5+cnFRUVUl5e\nLr6+vvLOO+/IrVu3ZOPGjeLk5CRz5swREZHVq1fLoEGDao1/8uRJERGZNm2ajBo1Sq5evSrFxcWS\nkJAgs2bNEhGRnTt3iqOjo8ydO1du3bolW7dulfbt20tRUZGIiDz//PMyZMgQKSwslKqqKtm7d6+U\nl5dLXl6eaDQaqaqqEhGR0aNHy7PPPiulpaXy888/y4ABA+Svf/2riIiMGzdOFixYICIi5eXl8vXX\nX9/19bq97oqKCgkICJCFCxdKZWWlZGVliVarlR9//FFERJ5++mnp1KmT7NmzR0REysrKao1XX12r\nV68WJycn+eijj6S6ulree+896dq1q3ldg8EgK1euvGt9d26/tLRU+vbtK/PmzZPKyko5deqU9OjR\nQ7744os6e/X19ZXc3FwREQkODpaAgAD54YcfzPMOHz4sIiKxsbHywgsvSHl5uRw+fFg8PT0lKytL\nRETmzp0rTk5OsmnTJhERuXnzpgwcOFCmT58uFRUVsmvXLtFqtTJhwoQ6a9i5c6fo9Xrzcz8/P/nq\nq69ERGTgwIGybt06ERG5ceOG7Nu3T0RETCaTxfee7IMB3wJ0795dOnbsKK6uruLq6iqPPfaYiNQE\nxc6dO83LPfvss+bA/kXPnj0lOztbsrOzLYJHRCQmJqZRAV9dXS0dOnQwh5KIyJ49e8Tf319Eav6D\nt2vXzuI/q5eXl+zfv1+qqqqkXbt28u2339bq6/aAP3/+vLRt21Zu3rxpnv/xxx/LkCFDRERk4sSJ\nMmXKFDlz5kyDr9ftAbpr1y7x9va2mJ+UlCSpqakiUhOwTz/99F3Haqiu1atXS2BgoHnejRs3RKPR\nyIULF0SkJuA/+uiju9Z35/b37dsnvr6+FssvWLBAkpOT66xvwoQJ8tZbb8m5c+ekZ8+eMnPmTHn/\n/ffl1KlT4urqKiIip0+fljZt2khJSYl5vVmzZskzzzwjIjUBHxcXZ56Xn58vjo6OUlpaap42fvx4\neeqpp+qsob6AHzx4sMydO1cuXrxosc6dv9zJPhzt/RcE1RzH3bRpEx566KFa824/eyE/Px9r1qyx\n+JO+srIS586dg4igW7duFut27969Udu/ePEiSktL0a9fP/M0EbE4Xuzh4QEHh/8d0Wvfvj1KSkpw\n6dIllJWVISAgoN5t5Ofno7KyEl26dDFPq66uhq+vLwBgyZIlmDNnDgYMGAA3NzdMnz4dycnJDdZe\nWFho8RoBNX0XFhYCqHlt9Xq91XUBsHg/pH379gCAkpISeHl5mbdxNxqNxuL7kp+fj8LCQotDL1VV\nVXd90zIuLg6bN2+GXq/H4MGDERcXh7Vr18LZ2dm8TmFhIdzd3dGhQwfzer6+vsjJyTE/v/01+GX7\n7dq1M0/r3r07CgoK7trH3axcuRKvv/46QkND4e/vj7lz5+K3v/1tk8eh5sGAb+FuDw9fX1/Mnj0b\nf/zjH2stl52djbNnz1pMy8/PR2BgIACgQ4cOKC0tNc+7/Sydzp07o127djh27JhF0DVG586d4ezs\njJ9++gm9e/e+63I+Pj5o27YtLl++bPGL4hc6nQ4ffPABAODrr7/G0KFDERcXhx49etS7/a5du6Kg\noAAiYn6t8vPzERIS0qj6G6qrIY15k/XO76G/vz+OHz/eqPHj4uIwY8YM6PV6GAwGDBo0CM8++yyc\nnZ0RFxcHoOY1uHLlCkpKStCxY0cANaeo3h7qt9fQpUsXXL16FaWlpeZfWPn5+WjTpk2jarpdYGAg\nPv74YwDAp59+isTERFy5ckXxM6fIOnyTtRWZPHky3n//fRw4cAAighs3buDf//43SkpKEBMTA0dH\nRyxfvhyVlZX47LPPcPDgQfO6kZGR+P7773HkyBGUlZUhNTXVPM/BwQGTJ0/GtGnTcPHiRQDA2bNn\nsX379gZrcnBwQEpKCl555RWcO3cOVVVV2Lt3LyoqKiyW69KlC4YNG4ZXXnkFxcXFqK6uxsmTJ83n\nU//jH//AmTNnANS8MafRaBoVuAMHDkT79u2xZMkSVFZWwmg0YsuWLeY3aaWBszgaqqshOp0OJ0+e\nvOv8O7c/YMAAaLVaLFmyBDdv3kRVVRW+++47i73t2wUGBsLZ2Rnr1q1DXFwctFotvLy88Omnn5oD\n3sfHBzExMZg1axbKy8vx7bffYtWqVXjqqafqHLN79+6Ijo7G3LlzUVlZif/85z/YsmVLo/q907p1\n68w/M506dTJ/3zw9PeHg4FDva0PNjwHfgt25F9SvXz98+OGHmDp1Ktzd3REUFIQ1a9YAAJycnPDZ\nZ58hPT0dHh4e+OSTTzBmzBhzwAQHB+P111/H0KFD0bNnT8TGxlqMv3jxYgQGBmLgwIHo1KkT4uPj\nLfYy69sje+ONNxAREYH+/fvDw8MDs2bNMm/39vXWrFmDiooKhIWFwd3dHY8//rj5L4mcnBwMHDgQ\nWq0Wo0aNwvLly+96LvntYzo5OeHzzz/Htm3b4OnpialTp2Lt2rUIDg42L9vQ3mR9ddW1/u3PX3rp\nJWzcuBHu7u6YNm1anbXevryDgwO2bNmCw4cPo0ePHvD09MSUKVNw/fr1u9ZnMBjQuXNn86Eeg8EA\nAOjbt695mfXr18NkMqFr164YM2YM0tLSzIf86urh448/xv79++Hu7o60tDQ8/fTT9b5Gd3sNv/ji\nC4SHh0Or1eLll1/Ghg0b0LZtW7Rv3x6zZ8/Ggw8+CDc3Nxw4cKDe8al5aKShXRxqtZKTk6HX6zFv\n3jx7l0JEdlDvHnxKSgp0Oh0iIiLM0w4cOIABAwYgKioK/fv3tzgMQC0Lf3cT/brVG/DJycnmizp+\n8dprr2HevHk4dOgQ0tLS8NprrzVrgWS9xhyeICL1qvcsmtjY2Fo3TerSpQuuXbsGACgqKqp1ah61\nHKtXr7Z3CURkRw0egzeZTEhISDBfEp+fn49BgwZBo9Gguroae/furXUeMvcaiYiso+Sh1SafRTNp\n0iQsX74cp0+fxttvv42UlJQ6l5Oaq2RV+W/u3Ll2r4H9sb9fW2+/hv6U1uSAP3DgAB577DEAQGJi\nIk9/IiJqoZoc8IGBgcjOzgYAZGVlmc83JiKilqXeN1mTkpKQnZ2NS5cuwcfHB2lpafjggw/wwgsv\noLy8HO3atTNfXv5r8suFJmrF/lovNfcGqL8/pTXLhU4ajaZZjicREamZ0tnJWxUQEakUA56ISKUY\n8EREKsWAJyJSKQY8EZFKMeCJiFSKAU9EZKPsbGDJEntXURsDnojIRn/9K/D55/auojYGPBGRjdq0\nAaZMsXcVtTHgiYhUigFPRKRSDHgiIpViwBMRqRQDnohIpRjwREQqxYAnIlKpegM+JSUFOp0OERER\nFtNXrFiB0NBQhIeHY+bMmc1aIBERWafej+xLTk7Giy++iIkTJ5qn7dy5E5s3b8a3334LJycnXLx4\nsdmLJCKipqs34GNjY2EymSymvffee5g1axacnJwAAJ6ennWum5qaan5sMBj4WYpEpFqlpdatZzQa\nYTQaFa3ldg1+JqvJZEJCQgKOHj0KAIiKisKoUaOQmZkJZ2dnvPHGG4iOjrYclJ/JSkS/IhoN8K9/\nAaNG2TqOstlZ7x58XW7duoWrV69i3759OHjwIJ544gmcOnVKsYKIiFobT0/ggQfsXUVtTT6LRq/X\nY8yYMQCA/v37w8HBAZcvX1a8MCIisk2TA3706NHIysoCABw/fhwVFRXw8PBQvDAiIrJNvYdokpKS\nkJ2djcuXL8PHxwdpaWlISUlBSkoKIiIicN9992HNmjX3qlYiImqCBt9ktWpQvslKRL8iXl7Ad9/V\nfLWF0tnJK1mJiFSKAU9EpFIMeCIilWLAExGpFAOeiEilGPBERCrFgCciUikGPBGRSjHgiYhUigFP\nRKRSDHgiIpViwBMRqRQDnohIpRjwREQqxYAnIlKpegM+JSUFOp0OERERtea9+eabcHBwwJUrV5qt\nOCIisl69AZ+cnIzMzMxa0wsKCvDll1+ie/fuzVYYERHZpt6Aj42NhZubW63pr7zyCpYsWdJsRRER\nke3q/UzWumzatAl6vR69e/eud7nU1FTzY4PBAIPB0NRNERG1aKWlQFYWUF5u3fpGoxFGo1HRmm7X\n4GeymkwmJCQk4OjRoygtLcWQIUPw5ZdfwsXFBf7+/sjJyYGHh4floPxMViL6Fdi8GZg8GYiLA9au\nBdq2tW08u34m68mTJ2EymRAZGQl/f3+cOXMG/fr1w88//6xYQURErYUIMHAg8Mkntod7c2jSIZqI\niAhcuHDB/Nzf3x/ffPMN3N3dFS+MiIhsU+8efFJSEmJiYnD8+HH4+Phg9erVFvM1Gk2zFkdERNar\ndw9+/fr19a586tQpRYshIiLl8EpWIiKVYsATEakUA56ISKUY8EREKsWAJyJSKQY8EZFKMeCJiFSK\nAU9EZKUXX7T+RmP3AgOeiMhKlZXAm2/au4q7Y8ATEVmpXbuafy0VA56ISKUY8EREKsWAJyJSKQY8\nEZFKMeCJiFSKAU9EpFL1BnxKSgp0Oh0iIiLM02bMmIHQ0FBERkZizJgxuHbtWrMXSURETVdvwCcn\nJyMzM9Ni2rBhw/D999/jyJEjCA4OxsKFC5u1QCIisk69AR8bGws3NzeLafHx8XBwqFnt/vvvx5kz\nZ5qvOiIislq9n8nakFWrViEpKanOeampqebHBoMBBoPBlk0REbUo5eVAXp5tYxiNRhiNRkXqqYtG\nRKS+BUwmExISEnD06FGL6fPnz0dubi4+/fTT2oNqNGhgWCKiVu3ECSA4uCbo77tPmTGVzk6r9uDT\n09OxdetWfPXVV4oVQkTU2gQGKhfuzaHJAZ+ZmYmlS5ciOzsbzs7OzVETEREpoN43WZOSkhATE4Mf\nf/wRPj4+WLVqFV588UWUlJQgPj4eUVFReP755+9VrURE1AQNHoO3alAegycilTtxAhgxouarUpTO\nTl7JSkSkUgx4IiKVYsATEakUA56ISKUY8EREKsWAJyJSKQY8EZEVSkrsXUHDGPBERFYYPBgICbF3\nFfXjhU5ERFZwdATKymq+KoUXOhERUaMw4ImIVIoBT0SkUgx4IiKVYsATEakUA56ISKUY8EREKlVv\nwKekpECn0yEiIsI87cqVK4iPj0dwcDCGDRuGoqKiZi+SiIiart6AT05ORmZmpsW0RYsWIT4+HseP\nH8fDDz+MRYsWNWuBRERknQavZDWZTEhISMDRo0cBACEhIcjOzoZOp8P58+dhMBjw3//+13JQXslK\nRCrXGq5kbXJpFy5cgE6nAwDodDpcuHChzuVSU1PNjw0GAwwGg1UFEhGpldFohNFobLbxm7wH7+bm\nhqtXr5rnu7u748qVK5aDcg+eiFSuNezBN/ksml8OzQDAuXPn4OXlpVgxRESknCYH/MiRI5GRkQEA\nyMjIwOjRoxUvioiIbFfvIZqkpCRkZ2fj0qVL0Ol0SEtLw6hRo/DEE0/g9OnT8PPzwyeffAJXV1fL\nQXmIhohUrjUcouH94ImIrNAaAp5XshIRqRQDnohIpRjwREQqxYAnIlIpBjwRkUox4ImIVIoBT0TU\nREeOANXV9q6iYTwPnoioifr0Abp3B/71L0CjUW5cngdPRGRn1dXAvHnKhntzYMATEakUA56ISKUY\n8EREKsWAJyJSKQY8EZFKMeCJiFSKAU9EpFJWB/zChQvRq1cvREREYPz48SgvL1eyLiIispFVAW8y\nmfDhhx8iNzcXR48eRVVVFTZs2KB0bUREZAOrPmzKxcUFTk5OKC0tRZs2bVBaWopu3bopXRsREdnA\nqoB3d3fH9OnT4evri3bt2mH48OEYOnSoxTKpqanmxwaDAQaDwZY6iYhahAcfBH74AXB2tn0so9EI\no9Fo+0B3YdXNxk6ePImEhATs3r0bnTp1wuOPP47ExEQ8+eSTNYPyZmNEpFKurkBODhAYqPzYLeJm\nYzk5OYiJiYGHhwccHR0xZswY7NmzR7GiiIhass6d7V1B41gV8CEhIdi3bx9u3rwJEcGOHTsQFham\ndG1ERGQDqwI+MjISEydORHR0NHr37g0AmDJliqKFERGRbfiBH0RETeDqCphMNV+V1iKOwRMRUcvH\ngCciUikGPBGRSjHgiYhUigFPRKRSDHgiIpWy6l40RES/Nv/5D7BjB1BWZu9KGo8BT0TUCOnpwM8/\nA/PmAS4u9q6mcRjwRESNNHIk8Lvf2buKxuMxeCIilWLAExGpFAOeiEilGPBERCrFgCciUikGPBGR\nSlkd8EVFRUhMTERoaCjCwsKwb98+JesiIiIbWX0e/EsvvYQRI0Zg48aNuHXrFm7cuKFkXUREZCOr\nAv7atWvYvXs3MjIyagZxdESnTp0ULYyIiGxjVcDn5eXB09MTycnJOHLkCPr164dly5ahffv25mVS\nU1PNjw0GAwwGg621EhHdcyLAhQtAcxykMBqNMBqNyg/8/6z6TNacnBw88MAD2LNnD/r3749p06bB\nxcUFaWlpNYPyM1mJSCWMRmD4cMDDA1i1CvjNb5pvWy3iM1n1ej30ej369+8PAEhMTERubq5iRRER\ntRQ3bwIPPQQUFjZvuDcHqwLe29sbPj4+OH78OABgx44d6NWrl6KFERGRbaw+i2bFihV48sknUVFR\ngYCAAKxevVrJuoiIyEZWB3xkZCQOHjyoZC1ERKQgXslKRKRSDHgiIpViwBMRqRQDnohIpRjwREQq\nxYAnIlIpBjwRkUpZfR48EZEavfEGkJ39v+c//wx07my/emzBgCcius2//w0MHQr07v2/acHB9qvH\nFgx4IqI7xMQAQ4bYuwrb8Rg8EZFKMeCJiFSKAU9EpFIMeCIilWLAExGpFAOeiEilrA74qqoqREVF\nISEhQcl6iIhIIVYH/LJlyxAWFgaNRqNkPUREpBCrAv7MmTPYunUrfve730FElK6JiIgUYNWVrC+/\n/DKWLl2K69ev33WZ1NRU82ODwQCDwWDNpoiIFPff/wInTtQ979Kle1eH0WiE0WhstvE10sRd8C1b\ntmDbtm34y1/+AqPRiDfffBOff/655aAaDffsiajF+s1vgKtXAS+v2vPatAGWLwd8fe99XUpnZ5P3\n4Pfs2YPNmzdj69atKCsrw/Xr1zFx4kSsWbNGsaKIiJqTCJCWBgwfbu9KmleT9+Bvl52djTfeeIN7\n8ETUqgwfDrzySssLeKWz0+bz4HkWDRFRy2TT7YLj4uIQFxenVC1ERKQgXslKRKRSDHgiIpViwBMR\nqRQDnohIpRjwREQqxYAnIlIpBjwRkUrZdB48EdG9dPAgMHp0za0GbHH5MjB7tjI1tWQMeCJqNc6f\nB8LCgIwM28ZxcAB0OmVqaskY8ETUqjg7A1272ruK1oHH4ImIVIoBT0SkUgx4IiKVYsATEakUA56I\nSKUY8EREKmVVwBcUFGDIkCHo1asXwsPDsXz5cqXrIiIiG1l1HryTkxPefvtt9OnTByUlJejXrx/i\n4+MRGhqqdH1ERGQlq/bgvb290adPHwBAx44dERoaisLCQkULIyIi29h8JavJZMKhQ4dw//33W0xP\nTU01PzYYDDAYDLZuiojuoZMngbVr7V2FpR9/tHcFyjIajTAajc02vkbE+tv2lJSUwGAw4E9/+hNG\njx79v0E1GtgwLBG1AO+/D3z0EfDoo/auxNIDDwDDh9u7iuahdHZavQdfWVmJsWPH4qmnnrIIdyJS\nj+ho4LY/xqmVseoYvIhg0qRJCAsLw7Rp05SuiYiIFGBVwH/99ddYt24ddu7ciaioKERFRSEzM1Pp\n2oiIyAZWHaIZNGgQqqurla6FiIgUxCtZiYhUigFPRKRSDHgiIpViwBMRqRQDnohIpRjwREQqZfO9\naO6G9x4jat0KCwGeDd26NVvAR0c318hEdC9cvAjMn2/vKsgWNt1s7K6D8mZjRERNpnR28hg8EZFK\nMeCJiFSKAU9EpFIMeCIilWLAExGpFAOeiEilGPBWaM4PyW0J2F/rpebeAPX3pzSrAz4zMxMhISEI\nCgrC4sWLlaypxVP7Dxn7a73U3Bug/v6UZlXAV1VVYerUqcjMzMSxY8ewfv16/PDDD0rXRkRENrAq\n4A8cOIDAwED4+fnByckJ48aNw6ZNm5SujYiIbGDVrQo2btyIL774Ah9++CEAYN26ddi/fz9WrFhR\nM6hGo2yVRES/EkreqsCqm401FOC8Dw0Rkf1ZdYimW7duKCgoMD8vKCiAXq9XrCgiIrKdVQEfHR2N\nEydOwGQyoaKiAn//+98xcuRIpWsjIiIbWHWIxtHREX/+858xfPhwVFVVYdKkSQgNDVW6NiIisoHV\n58E/8sgj+PHHH/HTTz9h1qxZ5ulqOD++oKAAQ4YMQa9evRAeHo7ly5cDAK5cuYL4+HgEBwdj2LBh\nKCoqMq+zcOFCBAUFISQkBNu3b7dX6Y1WVVWFqKgoJCQkAFBXb0VFRUhMTERoaCjCwsKwf/9+VfW3\ncOFC9OrVCxERERg/fjzKy8tbdX8pKSnQ6XSIiIgwT7Omn2+++QYREREICgrCSy+9dE97qE9d/c2Y\nMQOhoaGIjIzEmDFjcO3aNfM8RfsTBd26dUsCAgIkLy9PKioqJDIyUo4dO6bkJu6Jc+fOyaFDh0RE\npLi4WIKDg+XYsWMyY8YMWbx4sYiILFq0SGbOnCkiIt9//71ERkZKRUWF5OXlSUBAgFRVVdmt/sZ4\n8803Zfz48ZKQkCAioqreJk6cKCtXrhQRkcrKSikqKlJNf3l5eeLv7y9lZWUiIvLEE09Ienp6q+5v\n165dkpubK+Hh4eZpTemnurpaRET69+8v+/fvFxGRRx55RLZt23aPO6lbXf1t377d/H2YOXNms/Wn\naMDv2bNHhg8fbn6+cOFCWbhwoZKbsItRo0bJl19+KT179pTz58+LSM0vgZ49e4qIyIIFC2TRokXm\n5YcPHy579+61S62NUVBQIA8//LBkZWXJo48+KiKimt6KiorE39+/1nS19Hf58mUJDg6WK1euSGVl\npTz66KOyffv2Vt9fXl6eRQA2tZ/CwkIJCQkxT1+/fr38/ve/v0fVN+zO/m732WefyZNPPikiyven\n6L1ozp49Cx8fH/NzvV6Ps2fPKrmJe85kMuHQoUO4//77ceHCBeh0OgCATqfDhQsXAACFhYUWZxG1\n9L5ffvllLF26FA4O//v2q6W3vLw8eHp6Ijk5GX379sXkyZNx48YN1fTn7u6O6dOnw9fXF127doWr\nqyvi4+NV098vmtrPndO7devWKvoEgFWrVmHEiBEAlO9P0YBX2wVOJSUlGDt2LJYtWwatVmsxT6PR\n1NtvS30ttmzZAi8vL0RFRd31eoXW2hsA3Lp1C7m5uXj++eeRm5uLDh06YNGiRRbLtOb+Tp48iXfe\neQcmkwmFhYUoKSnBunXrLJZpzf3VpaF+WrP58+fjvvvuw/jx45tlfEUDXk3nx1dWVmLs2LGYMGEC\nRo8eDaBmT+L8+fMAgHPnzsHLywtA7b7PnDmDbt263fuiG2HPnj3YvHkz/P39kZSUhKysLEyYMEEV\nvQE1ezx6vR79+/cHACQmJiI3Nxfe3t6q6C8nJwcxMTHw8PCAo6MjxowZg71796qmv1805edRr9ej\nW7duOHPmjMX0lt5neno6tm7dir/97W/maUr3p2jAq+X8eBHBpEmTEBYWhmnTppmnjxw5EhkZGQCA\njIwMc/CPHDkSGzZsQEVFBfLy8nDixAkMGDDALrU3ZMGCBSgoKEBeXh42bNiAhx56CGvXrlVFbwDg\n7e0NHx9hAIOpAAABSElEQVQfHD9+HACwY8cO9OrVCwkJCaroLyQkBPv27cPNmzchItixYwfCwsJU\n098vmvrz6O3tDRcXF+zfvx8igrVr15rXaYkyMzOxdOlSbNq0Cc7Ozubpivdn/dsGddu6dasEBwdL\nQECALFiwQOnh74ndu3eLRqORyMhI6dOnj/Tp00e2bdsmly9flocffliCgoIkPj5erl69al5n/vz5\nEhAQID179pTMzEw7Vt94RqPRfBaNmno7fPiwREdHS+/eveWxxx6ToqIiVfW3ePFiCQsLk/DwcJk4\ncaJUVFS06v7GjRsnXbp0EScnJ9Hr9bJq1Sqr+snJyZHw8HAJCAiQF1980R6t1OnO/lauXCmBgYHi\n6+trzpfnnnvOvLyS/Vl1szEiImr5+IlOREQqxYAnIlIpBjwRkUox4ImIVIoBT0SkUgx4IiKV+j+i\nuqRWeAqilQAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x4bf2fd0>"
       ]
      }
     ],
     "prompt_number": 18
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Co-occurrence graph"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "An interesting question to ask is: which pairs of words co-occur in the same tweets?  We can find these relations and use them to construct a graph, which we can then analyze with NetworkX and plot with Matplotlib.\n",
      "\n",
      "We limit the graph to have at most `n_nodes` (for the most frequent words) just to keep the visualization easier to read."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "n_nodes = 10\n",
      "popular = sorted_wf[-n_nodes:]\n",
      "pop_words = [wc[0] for wc in popular]\n",
      "co_occur = tu.co_occurrences(lines, pop_words)\n",
      "wgraph = tu.co_occurrences_graph(popular, co_occur, cutoff=1)\n",
      "wgraph = nx.connected_component_subgraphs(wgraph)[0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 19
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "An interesting summary of the graph structure can be obtained by ranking nodes based on a centrality measure.  NetworkX offers several centrality measures, in this case we look at the [Eigenvector Centrality](http://networkx.lanl.gov/reference/generated/networkx.algorithms.centrality.eigenvector_centrality.html#networkx.algorithms.centrality.eigenvector_centrality):"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "centrality = nx.eigenvector_centrality_numpy(wgraph)\n",
      "tu.summarize_centrality(centrality)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "Graph centrality\n",
        "         google: 0.424\n",
        "       ventures: 0.421\n",
        "          funds: 0.421\n",
        "         tracks: 0.421\n",
        "      discovery: 0.421\n",
        "           like: 0.33\n",
        "        obama's: 0.0398\n"
       ]
      }
     ],
     "prompt_number": 20
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "And we can use this measure to provide an interesting view of the structure of our query dataset:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print \"Graph visualization for query:\", query\n",
      "tu.plot_graph(wgraph, tu.centrality_layout(wgraph, centrality), plt.figure(figsize=(8,8)),\n",
      "    title='Centrality and term co-occurrence graph, q=\"%s\"' % query)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Graph visualization for query: Big Data\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAk4AAAInCAYAAABuotZnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x4f15310>"
       ]
      }
     ],
     "prompt_number": 21
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}