{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "754bb952",
   "metadata": {},
   "source": [
    "# Parts of Speech Tagging"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06edab3f",
   "metadata": {},
   "source": [
    "POS tagging is a common processing step in NLP.\n",
    "\n",
    "If we have POS tagged, we can often already determine who is doing what to whom:\n",
    "\n",
    "- N V N\n",
    "- Subject Action Object\n",
    "- John eats pizza\n",
    "\n",
    "But this is kind of illusory, because \"correct\" tagging actually requires correct parsing:\n",
    "\n",
    "- Fruit flies like banana.\n",
    "- N V PP N\n",
    "- N N V N\n",
    "\n",
    "\n",
    "\"Time flies like an arrow, but fruit flies like banana.\"\n",
    "\n",
    "- Buffalo buffalo buffalo buffalo.\n",
    "- A N V N\n",
    "- N V A N\n",
    "- A A N V\n",
    "- V A A N\n",
    "\n",
    "Still, we can tag correctly often enough using statistical methods to be practically useful for information retrieval. However, POS tagging can never be totally correct, it's always a heuristic.\n",
    "\n",
    "(We could require a POS tagger to return all possible tags for each word, or all possible overall assignments for tags.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5f7e85d5",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from nltk.corpus import wordnet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cb0104ee",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " american_bison.n.01 large shaggy-haired brown bison of North American plains\n",
      "buffalo.n.02 a city on Lake Erie in western New York (near Niagara Falls)\n",
      "buffalo.n.03 meat from an American bison\n",
      "old_world_buffalo.n.01 any of several Old World animals resembling oxen including, e.g., water buffalo; Cape buffalo\n",
      "buffalo.v.01 intimidate or overawe\n"
     ]
    }
   ],
   "source": [
    "for s in wordnet.synsets(\"buffalo\"):\n",
    "    print s.name,s.definition"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50fbd2f5",
   "metadata": {},
   "source": [
    "In part, we're looking at how NLTK does this because in terms of\n",
    "research, it's interesting to ask how we can do better (say,\n",
    "using recurrent neural networks or finite state transducers)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beca34a2",
   "metadata": {},
   "source": [
    "# Tagged Corpora"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "1d42443c",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from nltk.corpus import brown\n",
    "import nltk.tag # fix the default simplification map\n",
    "nltk.tag.brown_mapping1[\"dt\"] = \"DET\"\n",
    "nltk.tag.brown_mapping1[\"do\"] = \"V\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "84cfc7ef",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "tagged = brown.tagged_words()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "cd15b312",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1161192\n",
      "[('The', 'AT'), ('Fulton', 'NP-TL'), ('County', 'NN-TL'), ('Grand', 'JJ-TL'), ('Jury', 'NN-TL'), ('said', 'VBD'), ('Friday', 'NR'), ('an', 'AT'), ('investigation', 'NN'), ('of', 'IN'), (\"Atlanta's\", 'NP$'), ('recent', 'JJ'), ('primary', 'NN'), ('election', 'NN'), ('produced', 'VBD'), ('``', '``'), ('no', 'AT'), ('evidence', 'NN'), (\"''\", \"''\"), ('that', 'CS'), ('any', 'DTI'), ('irregularities', 'NNS'), ('took', 'VBD'), ('place', 'NN'), ('.', '.'), ('The', 'AT'), ('jury', 'NN'), ('further', 'RBR'), ('said', 'VBD'), ('in', 'IN'), ('term-end', 'NN'), ('presentments', 'NNS'), ('that', 'CS'), ('the', 'AT'), ('City', 'NN-TL'), ('Executive', 'JJ-TL'), ('Committee', 'NN-TL'), (',', ','), ('which', 'WDT'), ('had', 'HVD')]\n"
     ]
    }
   ],
   "source": [
    "print len(tagged)\n",
    "print tagged[:40]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "4079d0c9",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1161192\n",
      "[('The', 'DET'), ('Fulton', 'NP'), ('County', 'N'), ('Grand', 'ADJ'), ('Jury', 'N'), ('said', 'VD'), ('Friday', 'N'), ('an', 'DET'), ('investigation', 'N'), ('of', 'P'), (\"Atlanta's\", 'NP'), ('recent', 'ADJ'), ('primary', 'N'), ('election', 'N'), ('produced', 'VD'), ('``', '``'), ('no', 'DET'), ('evidence', 'N'), (\"''\", \"''\"), ('that', 'CNJ'), ('any', 'DET'), ('irregularities', 'N'), ('took', 'VD'), ('place', 'N'), ('.', '.'), ('The', 'DET'), ('jury', 'N'), ('further', 'ADV'), ('said', 'VD'), ('in', 'P'), ('term-end', 'N'), ('presentments', 'N'), ('that', 'CNJ'), ('the', 'DET'), ('City', 'N'), ('Executive', 'ADJ'), ('Committee', 'N'), (',', ','), ('which', 'WH'), ('had', 'V')]\n"
     ]
    }
   ],
   "source": [
    "tagged = brown.tagged_words(simplify_tags=True)\n",
    "print len(tagged)\n",
    "print tagged[:40]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be757880",
   "metadata": {},
   "source": [
    "Most common proper nouns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "08c2035d",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9765\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['Blaber',\n",
       " 'Black',\n",
       " \"Black's\",\n",
       " 'Blackberry',\n",
       " 'Blackfeet',\n",
       " 'Blackman',\n",
       " 'Blackmer',\n",
       " 'Blacks',\n",
       " 'Blackstone',\n",
       " 'Blackwell']"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np = sorted(list(set([n for n,c in tagged if c==\"NP\"])))\n",
    "print len(np)\n",
    "np[1000:1010]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "990cafcd",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "ADJ\tadjective\tnew, good, high, special, big, local\n",
    "ADV\tadverb\treally, already, still, early, now\n",
    "CNJ\tconjunction\tand, or, but, if, while, although\n",
    "DET\tdeterminer\tthe, a, some, most, every, no\n",
    "EX\t existential\tthere, there's\n",
    "FW\t foreign word\tdolce, ersatz, esprit, quo, maitre\n",
    "MOD\tmodal verb\twill, can, would, may, must, should\n",
    "N\t  noun\tyear, home, costs, time, education\n",
    "NP\t proper noun\tAlison, Africa, April, Washington\n",
    "NUM\tnumber\ttwenty-four, fourth, 1991, 14:24\n",
    "PRO\tpronoun\the, their, her, its, my, I, us\n",
    "P\t  preposition\ton, of, at, with, by, into, under\n",
    "TO\t the word to\tto\n",
    "UH\t interjection\tah, bang, ha, whee, hmpf, oops\n",
    "V\t  verb\tis, has, get, do, make, see, run\n",
    "VD\t past tense\tsaid, took, told, made, asked\n",
    "VG     present participle\tmaking, going, playing, working\n",
    "VN\t past participle\tgiven, taken, begun, sung\n",
    "WH\t wh determiner\twho, which, when, what, where, how\n",
    "\"\"\"; None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35f08e21",
   "metadata": {},
   "source": [
    "Most common verbs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "1b9e9ad7",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(('is', 'V'), 10010),\n",
       " (('was', 'V'), 9777),\n",
       " (('be', 'V'), 6342),\n",
       " (('had', 'V'), 5101),\n",
       " (('are', 'V'), 4333),\n",
       " (('have', 'V'), 3892),\n",
       " (('were', 'V'), 3279),\n",
       " (('been', 'V'), 2470),\n",
       " (('has', 'V'), 2424),\n",
       " (('said', 'VD'), 1738)]"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Counter((w,t) for w,t in tagged if t.startswith(\"V\")).most_common(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "1eb7390b",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "freq = nltk.FreqDist(tag for word,tag in tagged)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "9662de50",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('N', 229788),\n",
       " ('DET', 136425),\n",
       " ('P', 122591),\n",
       " ('V', 87519),\n",
       " ('ADJ', 71994),\n",
       " ('PRO', 66793),\n",
       " ('.', 61254),\n",
       " ('CNJ', 60326),\n",
       " (',', 58336),\n",
       " ('ADV', 43934)]"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freq.items()[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "3c5ffd59",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEvCAYAAAB49NeYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "freq.plot(cumulative=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8fd67669",
   "metadata": {},
   "source": [
    "# N-Grams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "951eb706",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Counter({'N': 229788, 'DET': 136425, 'P': 122591, 'V': 87519, 'ADJ': 71994, 'PRO': 66793, '.': 61254, 'CNJ': 60326, ',': 58336, 'ADV': 43934, 'NP': 43179, 'VN': 29932, 'VD': 26192, 'VG': 18188, 'NUM': 17024, 'TO': 14998, 'WH': 14919, 'MOD': 13345, '``': 8837, \"''\": 8789, 'VBZ': 7469, '*': 4613, '': 3431, ')': 2457, '(': 2426, 'EX': 2280, ':': 1718, 'FW': 1226, 'UH': 629, \"'\": 317, 'NIL': 157, 'VB+PPO': 71, 'VBG+TO': 17, 'VBN+TO': 5, 'VB+TO': 4, 'VB+IN': 3, 'VB+AT': 2, 'VB+RP': 2, 'VB+VB': 1, 'VB+JJ': 1})"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from collections import Counter\n",
    "Counter(c for w,c in tagged)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "7c3e0324",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(('DET', 'N'), 686), (('N', 'P'), 603), (('N', 'N'), 452), (('P', 'DET'), 448), (('ADJ', 'N'), 412), (('P', 'N'), 238), (('N', '.'), 235), (('DET', 'ADJ'), 216), (('N', ','), 204), (('V', 'DET'), 159)]\n"
     ]
    }
   ],
   "source": [
    "bg = nltk.bigrams(tag for word,tag in tagged[:10000])\n",
    "freq = nltk.FreqDist(bg)\n",
    "print freq.items()[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "id": "d247fde6",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from itertools import *\n",
    "def ibigrams(l):\n",
    "    return izip(l,islice(l,1,None))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "cb4c8f64",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def gcat(l):\n",
    "    return \" \".join([x[1] for x in l])\n",
    "def gwords(l):\n",
    "    return \" \".join([x[0] for x in l])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "ab99eaa0",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'DET NP'"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gcat(nltk.bigrams(tagged[:10])[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "a2de2471",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('DET N', 74300),\n",
       " ('N P', 59944),\n",
       " ('P DET', 52654),\n",
       " ('ADJ N', 45836),\n",
       " ('DET ADJ', 29884),\n",
       " ('N .', 29522),\n",
       " ('N ,', 28025),\n",
       " ('P N', 25040),\n",
       " ('N N', 24310),\n",
       " ('N CNJ', 18705)]"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bic = Counter(gcat(x) for x in ibigrams(tagged))\n",
    "bic.most_common(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "93323954",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['PRO N VD ADJ N .', 'NP , NP ( N )', 'NP , NP ( N )']"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sent5 = [gcat(s) for s in brown.tagged_sents(simplify_tags=True) if len(s)==6]\n",
    "sent5[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "28d9bd53",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[((('the', 'DET'), ('U.S.', 'NP')), 76),\n",
       " ((('the', 'DET'), ('Congo', 'NP')), 49),\n",
       " ((('the', 'DET'), ('U.N.', 'NP')), 37),\n",
       " ((('the', 'DET'), ('U.', 'NP')), 36),\n",
       " ((('the', 'DET'), ('Bible', 'NP')), 34),\n",
       " ((('the', 'DET'), ('Negro', 'NP')), 31),\n",
       " ((('the', 'DET'), ('Congress', 'NP')), 27),\n",
       " ((('the', 'DET'), ('Russians', 'NP')), 23),\n",
       " ((('the', 'DET'), ('Chicago', 'NP')), 23),\n",
       " ((('the', 'DET'), ('Lo', 'NP')), 19)]"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Counter(x for x in ibigrams(tagged) if gcat(x)==\"DET NP\").most_common(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "4fbb5c40",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('American Catholic higher education', 4),\n",
       " ('light green nonspecific fluorescence', 2),\n",
       " ('early English historical circles', 1),\n",
       " ('plain old black dirt', 1),\n",
       " ('Such additional daytime class', 1),\n",
       " ('sturdy little three-masted bark', 1),\n",
       " ('such traditional Japanese foods', 1),\n",
       " ('official American Catholic project', 1),\n",
       " ('hard cruel bony skull', 1),\n",
       " ('continuous critical public scrutiny', 1)]"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Counter(gwords(x) for x in nltk.ingrams(tagged,4) if gcat(x)==\"ADJ ADJ ADJ N\").most_common(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e455ab4",
   "metadata": {},
   "source": [
    "# Relationship Mining"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dbd2383e",
   "metadata": {},
   "source": [
    "For relationship mining, we try to determine how related two objects\n",
    "are by seeing how often they are mentioned together.\n",
    "\n",
    "There are many ways in which this can be done.\n",
    "For example, relationships may be given explicitly in Facebook or FOAF.\n",
    "\n",
    "But often, we only have text available.\n",
    "Let's try to do some simple relationship mining on the Brown corpus.\n",
    "\n",
    "We start with the tagged sentences, and assume that there is some\n",
    "relationship between two objects if they are mentioned in the same sentence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "id": "655b3099",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('The', 'DET'),\n",
       " ('September-October', 'NP'),\n",
       " ('term', 'N'),\n",
       " ('jury', 'N'),\n",
       " ('had', 'V'),\n",
       " ('been', 'V'),\n",
       " ('charged', 'VN'),\n",
       " ('by', 'P'),\n",
       " ('Fulton', 'NP'),\n",
       " ('Superior', 'ADJ'),\n",
       " ('Court', 'N'),\n",
       " ('Judge', 'N'),\n",
       " ('Durwood', 'NP'),\n",
       " ('Pye', 'NP'),\n",
       " ('to', 'TO'),\n",
       " ('investigate', 'V'),\n",
       " ('reports', 'N'),\n",
       " ('of', 'P'),\n",
       " ('possible', 'ADJ'),\n",
       " ('``', '``'),\n",
       " ('irregularities', 'N'),\n",
       " (\"''\", \"''\"),\n",
       " ('in', 'P'),\n",
       " ('the', 'DET'),\n",
       " ('hard-fought', 'ADJ'),\n",
       " ('primary', 'N'),\n",
       " ('which', 'WH'),\n",
       " ('was', 'V'),\n",
       " ('won', 'VN'),\n",
       " ('by', 'P'),\n",
       " ('Mayor-nominate', 'N'),\n",
       " ('Ivan', 'NP'),\n",
       " ('Allen', 'NP'),\n",
       " ('Jr.', 'NP'),\n",
       " ('.', '.')]"
      ]
     },
     "execution_count": 196,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stagged = brown.tagged_sents(simplify_tags=True)\n",
    "stagged[2]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eaecff01",
   "metadata": {},
   "source": [
    "Tagging is on a per-word basis, but \"objects\" are often referred to by multiple names.\n",
    "So, we need to write a function to collapse multiple adjacent words if they have\n",
    "the same tag.\n",
    "\n",
    "The `itertools.groupby` function helps us with that.\n",
    "You should learn about `itertools`; it is somewhat analogous\n",
    "to UNIX command line programs like `uniq` etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "id": "7abc3e09",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('The', 'DET'),\n",
       " ('September-October', 'NP'),\n",
       " ('term jury', 'N'),\n",
       " ('had been', 'V'),\n",
       " ('charged', 'VN'),\n",
       " ('by', 'P'),\n",
       " ('Fulton', 'NP'),\n",
       " ('Superior', 'ADJ'),\n",
       " ('Court Judge', 'N'),\n",
       " ('Durwood Pye', 'NP'),\n",
       " ('to', 'TO'),\n",
       " ('investigate', 'V'),\n",
       " ('reports', 'N'),\n",
       " ('of', 'P'),\n",
       " ('possible', 'ADJ'),\n",
       " ('``', '``'),\n",
       " ('irregularities', 'N'),\n",
       " (\"''\", \"''\"),\n",
       " ('in', 'P'),\n",
       " ('the', 'DET'),\n",
       " ('hard-fought', 'ADJ'),\n",
       " ('primary', 'N'),\n",
       " ('which', 'WH'),\n",
       " ('was', 'V'),\n",
       " ('won', 'VN'),\n",
       " ('by', 'P'),\n",
       " ('Mayor-nominate', 'N'),\n",
       " ('Ivan Allen Jr.', 'NP'),\n",
       " ('.', '.')]"
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def collapse(s):\n",
    "    \"\"\"Collapse adjacent words if the have same tag.\"\"\"\n",
    "    return [(\" \".join(x[0] for x in g),k) for k,g in groupby(s,lambda x:x[1])]\n",
    "collapse(stagged[2])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b08c208",
   "metadata": {},
   "source": [
    "Next, for each sentence, we just filter out the proper nouns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "id": "f30e7038",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['September-October', 'Fulton', 'Durwood Pye', 'Ivan Allen Jr.']"
      ]
     },
     "execution_count": 245,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def NPlist(s):\n",
    "    \"\"\"Return a list of proper nouns in a sequence of tags.\"\"\"\n",
    "    return [x[0] for x in s if x[1]==\"NP\"]\n",
    "\n",
    "NPlist(collapse(stagged[2]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "id": "657e3470",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['Fulton', \"Atlanta's\"],\n",
       " ['Atlanta'],\n",
       " ['September-October', 'Fulton', 'Durwood Pye', 'Ivan Allen Jr.'],\n",
       " [],\n",
       " [\"Georgia's\"],\n",
       " ['Fulton'],\n",
       " ['Atlanta', 'Fulton'],\n",
       " [],\n",
       " [],\n",
       " []]"
      ]
     },
     "execution_count": 246,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nps = (NPlist(collapse(s)) for s in stagged)\n",
    "list(islice(nps,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a65b303",
   "metadata": {},
   "source": [
    "We really just want those sentences that actually mention more than one proper noun."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "id": "0f4c908c",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['Fulton', 'Fulton'],\n",
       " [\"Atlanta's\", 'Jan.'],\n",
       " ['Atlanta', 'Fulton', 'Bellwood', 'Alpharetta', 'Grady', 'Fulton'],\n",
       " ['William B. Hartsfield', 'Pearl Williams Hartsfield', 'Fulton'],\n",
       " ['William Berry Jr.', 'Mrs. J. M. Cheshire', 'Griffin'],\n",
       " ['Opelika', 'Ala.'],\n",
       " ['Hartsfield', 'E. Pelham'],\n",
       " ['Hartsfield', 'Atlanta'],\n",
       " ['Ivan Allen Jr.', 'Sept.', 'Hartsfield'],\n",
       " ['James W. Dorsey', 'Sept.', 'Savannah', 'Texas', 'John Tower']]"
      ]
     },
     "execution_count": 247,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nps = (x for x in nps if len(x)>1)\n",
    "list(islice(nps,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ec4c83b",
   "metadata": {},
   "source": [
    "As you can see, there is a wide variety of ways of referring to people,\n",
    "including titles, initials, etc.\n",
    "\n",
    "Solving the problem of mapping names to identities in general is a hard\n",
    "problem (c.f. authorship attribution).\n",
    "But we can prototype something with a little bit of regular expression magic."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "id": "9dfcb92b",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Kennedy'"
      ]
     },
     "execution_count": 259,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def NPnormalize(s,targets=[\"Kennedy\",\"Khrushchev\"]):\n",
    "    \"\"\"Normalize proper names by stripping prefixes/suffixes etc.\"\"\"\n",
    "    for t in targets: \n",
    "        if t in s: return t\n",
    "    s = re.sub(r'^(Mr|Mrs|Dr|Prof|Professor)\\.?','',s)\n",
    "    s = re.sub(r'[A-Z]\\.\\s','',s)\n",
    "    s = re.sub(r\"'s\",\"\",s)\n",
    "    s = re.sub(r'\\W+',' ',s)\n",
    "    return s.strip()\n",
    "NPnormalize(\"Mr. Kennedy, Jr.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b13c16a9",
   "metadata": {},
   "source": [
    "With these functions, we can now get a reasonable list of proper nouns\n",
    "for each sentence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "id": "6c0c8305",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['Sam Caldwell', 'Garland Byrd'],\n",
       " ['Rob Ledford', 'Gainesville'],\n",
       " ['Caldwell', 'Byrd'],\n",
       " ['Vandiver', 'Marvin Griffin'],\n",
       " ['Schley', 'B D Pelham'],\n",
       " ['Pelham', 'Georgia', 'November'],\n",
       " ['Jackson', 'Georgia'],\n",
       " ['Barber', 'Washington'],\n",
       " ['Harry Davis', 'Felix Bush', 'Miller'],\n",
       " ['Davis', 'Bush']]"
      ]
     },
     "execution_count": 260,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nps = [[NPnormalize(x) for x in l] for l in nps]\n",
    "list(islice(nps,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59a71b4e",
   "metadata": {},
   "source": [
    "We are interested in co-occurring pairs. Instead of nested loops,\n",
    "we use `itertools.combinations` to find these."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "id": "e5699653",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(('Brown', 'Sharpe'), 35),\n",
       " (('D C', 'Washington'), 31),\n",
       " (('B', 'C'), 29),\n",
       " (('God', 'God'), 26),\n",
       " (('Mantle', 'Maris'), 23),\n",
       " (('B', 'O'), 22),\n",
       " (('Paris', 'Paris'), 22),\n",
       " (('A', 'B'), 21),\n",
       " (('Palmer', 'Player'), 19),\n",
       " (('Pike', 'Woodruff'), 18)]"
      ]
     },
     "execution_count": 261,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pairs = [tuple(sorted(x)) for l in nps for x in combinations(l,2)]\n",
    "Counter(pairs).most_common(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf5f878c",
   "metadata": {},
   "source": [
    "There are still some pairs in here that we don't want. Let's filter those out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "id": "08305822",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(('Brown', 'Sharpe'), 35),\n",
       " (('D C', 'Washington'), 31),\n",
       " (('Mantle', 'Maris'), 23),\n",
       " (('Palmer', 'Player'), 19),\n",
       " (('Pike', 'Woodruff'), 18),\n",
       " (('Oct', 'Sept'), 17),\n",
       " (('Delaware', 'Du Pont'), 16),\n",
       " (('God', 'Jesus Christ'), 16),\n",
       " (('Miriam', 'Wright'), 16),\n",
       " (('Christiana', 'Delaware'), 15)]"
      ]
     },
     "execution_count": 262,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pairs = [(u,v) for u,v in pairs if len(u)>2 and len(v)>2 and u!=v]\n",
    "Counter(pairs).most_common(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01104f52",
   "metadata": {},
   "source": [
    "The entire graph is too large, but let's focus on two nodes: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "id": "d20bcd0d",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13\n",
      "[(('Congress', 'Khrushchev'), 7), (('Eisenhower', 'Kennedy'), 4), (('Kennedy', 'Laos'), 6), (('Kennedy', 'U S'), 4), (('Khrushchev', 'U S'), 6), (('Khrushchev', 'Moscow'), 4), (('China', 'Khrushchev'), 5), (('Cuba', 'Kennedy'), 4), (('Kennedy', 'Khrushchev'), 8), (('Kennedy', 'Nixon'), 4)]\n"
     ]
    }
   ],
   "source": [
    "relations = [p for p in Counter(pairs).items() \\\n",
    "    if re.search(\"Kennedy|Khrushchev\",\" \".join(p[0])) and p[1]>3]\n",
    "print len(relations)\n",
    "print relations[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "id": "c1247624",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0xee7e750>"
      ]
     },
     "execution_count": 264,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqsAAAHHCAYAAACYxRFNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pydot\n",
    "graph = pydot.Dot()\n",
    "for (u,v),n in relations:\n",
    "    graph.add_edge(pydot.Edge(u,v,label=str(n)))\n",
    "graph.write_png(\"/tmp/_test.png\",prog=\"fdp\")\n",
    "figsize(12,12)\n",
    "gca().set_xticks([]); gca().set_yticks([])\n",
    "imshow(imread(\"/tmp/_test.png\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9139a9c",
   "metadata": {},
   "source": [
    "For visualizing large relationship graphs, there are special tools:\n",
    "\n",
    "- http://gephi.org/\n",
    "- http://gephi.org/2010/openord-new-layout-plugin-the-fastest-algorithm-so-far/\n",
    "- http://www.textarc.org/\n",
    "- ..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60b8c757",
   "metadata": {},
   "source": [
    "# Automatic Tagging"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc289dca",
   "metadata": {},
   "source": [
    "Although the above results are neat, they aren't all that useful in practice\n",
    "because most texts we want to visualize in such ways aren't tagged, and tagging\n",
    "them by hand ist costly.\n",
    "\n",
    "What we need is an *automated tagger*."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff38912a",
   "metadata": {},
   "source": [
    "Let's take a page off Wikipedia and tag it automatically."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "id": "532adc73",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'<!DOCTYPE html>\\n<html lang=\"en\" dir=\"ltr\" class=\"client-nojs\">\\n<head>\\n<title>George Washington - Wik'"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "opener = urllib2.build_opener()\n",
    "opener.addheaders = [('User-agent', 'Mozilla/5.0')]\n",
    "infile = opener.open('http://en.wikipedia.org/w/index.php?title=George_Washington&printable=yes')\n",
    "page = infile.read().decode(\"utf-8\")\n",
    "page[:400]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "id": "31478278",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "George Washington - Wikipedia, the free encyclopedia\n",
      "\t\t\t\t\t\t\t\t\t\t\t\t George Washington\n",
      "\t\t\t\t\t\t\t\t\t\t\t\t From Wikipedia, the free encyclopedia\n",
      "\t\t\t\t\tJump to:\t\t\t\t\t navigation , \t\t\t\t\t search\n",
      "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t This article is about the first President of the United States. For other uses, see George Washington (disambiguation) .\n",
      " For a simpler version of this article, see the Simple English Wikipedia article: \n"
     ]
    }
   ],
   "source": [
    "page = nltk.util.clean_html(page)\n",
    "page = re.sub(r'\\s*\\n+','\\n',page)\n",
    "print page[:400]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "id": "8c134dd1",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(u'Washington', 'NNP'),\n",
       " (u'quickly', 'RB'),\n",
       " (u'became', 'VBD'),\n",
       " (u'a', 'DT'),\n",
       " (u'senior', 'JJ'),\n",
       " (u'officer', 'NN'),\n",
       " (u'in', 'IN'),\n",
       " (u'the', 'DT'),\n",
       " (u'colonial', 'JJ'),\n",
       " (u'forces', 'NNS'),\n",
       " (u'during', 'IN'),\n",
       " (u'the', 'DT'),\n",
       " (u'first', 'JJ'),\n",
       " (u'stages', 'NNS'),\n",
       " (u'of', 'IN'),\n",
       " (u'the', 'DT'),\n",
       " (u'French', 'JJ'),\n",
       " (u'and', 'CC'),\n",
       " (u'Indian', 'JJ'),\n",
       " (u'War', 'NN'),\n",
       " (u'.', '.')]"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sents = nltk.sent_tokenize(page)\n",
    "sents = [nltk.word_tokenize(s) for s in sents]\n",
    "stagged = [nltk.pos_tag(s) for s in sents]\n",
    "stagged[17]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "id": "48cbe580",
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(u'Washington', 'N'),\n",
       " (u'quickly', 'ADV'),\n",
       " (u'became', 'VD'),\n",
       " (u'a', 'DET'),\n",
       " (u'senior', 'ADJ'),\n",
       " (u'officer', 'N'),\n",
       " (u'in', 'P'),\n",
       " (u'the', 'DET'),\n",
       " (u'colonial', 'ADJ'),\n",
       " (u'forces', 'N'),\n",
       " (u'during', 'P'),\n",
       " (u'the', 'DET'),\n",
       " (u'first', 'ADJ'),\n",
       " (u'stages', 'N'),\n",
       " (u'of', 'P'),\n",
       " (u'the', 'DET'),\n",
       " (u'French', 'ADJ'),\n",
       " (u'and', 'CNJ'),\n",
       " (u'Indian', 'ADJ'),\n",
       " (u'War', 'N'),\n",
       " (u'.', '.')]"
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def simplify_tags(l):\n",
    "    return [(w,nltk.tag.simplify_brown_tag(t)) for w,t in l]\n",
    "stagged = [simplify_tags(l) for l in stagged]\n",
    "stagged[17]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "01ddb8e8",
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {},
 "nbformat": 4,
 "nbformat_minor": 5
}