IST664_HW1_Hank.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "IST664-HW1-Hank.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "TwCcd92lxlzV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Week 2:  Bigram Frequencies and Mutual Information\n",
        "# This file has small examples that are meant to be run individually\n",
        "# in the Python interpreter or jupyter notebook cells"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J-l_9rHbGrMH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import pandas as pd\n",
        "import io"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ztELZ6inx9Wq",
        "colab_type": "code",
        "outputId": "3cb8a6a0-6fc4-4220-978d-ce2b379918c9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 104
        }
      },
      "source": [
        "# Getting started to process a text example\n",
        "import nltk\n",
        "from nltk import FreqDist\n",
        "nltk.download('punkt')\n",
        "nltk.download('stopwords')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Package punkt is already up-to-date!\n",
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VdDA_CuByADY",
        "colab_type": "code",
        "outputId": "55f48e72-6a4a-42d8-e1aa-eafefd02527d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# get the text of the lyrics, tokenize it,\n",
        "#   and reduce the tokens to lowercase.\n",
        "#file0 = nltk.corpus.gutenberg.fileids( ) [3]\n",
        "#hanktext = nltk.corpus.gutenberg.raw(file0)\n",
        "f = open('hankedit.txt')\n",
        "raw = f.read()\n",
        "hanktokens = nltk.word_tokenize(raw) \n",
        "hankwords = [w.lower( ) for w in hanktokens] \n",
        "# show some of thv words\n",
        "print(len(hankwords))\n",
        "print(hankwords[ :110])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "8050\n",
            "['if', 'the', 'wife', 'and', 'i', 'are', 'fussin', \"'\", ',', 'brother', ',', 'that', \"'s\", 'our', 'right', \"'cause\", 'me', 'and', 'that', 'sweet', 'woman', \"'s\", 'got', 'a', 'license', 'to', 'fight', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', ',', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', \"'\", 'mine', 'oh', ',', 'the', 'woman', 'on', 'our', 'party', 'line', \"'s\", 'the', 'nosiest', 'thing', 'she', 'picks', 'up', 'her', 'receiver', 'when', 'she', 'knows', 'it', \"'s\", 'my', 'ring', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', ',', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', \"'\", 'mine', 'i', 'got', 'a']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gj_3iCE9yEH8",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Creating a frequency distribution of words\n",
        "ndist = FreqDist(hankwords)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "98JgMj83yGmG",
        "colab_type": "code",
        "outputId": "d25aac3e-1324-4620-97c0-623ecdd5ce8e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        }
      },
      "source": [
        "# print the top 50 tokens by frequency\n",
        "nitems = ndist.most_common(50)\n",
        "for item in nitems:\n",
        "    print (item[0], '\\t', item[1])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "i \t 327\n",
            "the \t 308\n",
            ", \t 234\n",
            "and \t 215\n",
            "a \t 190\n",
            "you \t 184\n",
            "my \t 143\n",
            "me \t 133\n",
            "to \t 124\n",
            "in \t 116\n",
            "'s \t 115\n",
            "on \t 110\n",
            "' \t 104\n",
            "'' \t 104\n",
            "'ll \t 100\n",
            "it \t 90\n",
            "of \t 83\n",
            "that \t 74\n",
            "your \t 74\n",
            "n't \t 71\n",
            "got \t 56\n",
            "he \t 56\n",
            "do \t 53\n",
            "'m \t 52\n",
            "lord \t 48\n",
            "when \t 47\n",
            "she \t 45\n",
            "we \t 45\n",
            "love \t 44\n",
            "but \t 44\n",
            "no \t 44\n",
            "have \t 44\n",
            "be \t 43\n",
            "all \t 43\n",
            "for \t 42\n",
            "over \t 42\n",
            "so \t 40\n",
            "is \t 40\n",
            "heart \t 36\n",
            "was \t 36\n",
            "with \t 35\n",
            "there \t 32\n",
            "oh \t 30\n",
            "like \t 30\n",
            "- \t 29\n",
            "just \t 28\n",
            "down \t 28\n",
            "will \t 27\n",
            "if \t 26\n",
            "new \t 26\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V3cIL9_pviW7",
        "colab_type": "code",
        "outputId": "b3511a81-6a9c-4d43-8e03-423f6f3187b1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        }
      },
      "source": [
        "ndist.plot(30,cumulative=False)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEbCAYAAADAsRPLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSB3Q0NCulQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3hnK8NpmyNnQ",
        "colab_type": "code",
        "outputId": "1d364d4c-c51a-42c2-8da9-0e5562e6003e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 55
        }
      },
      "source": [
        "print(hankwords[:160])\n",
        "#print(hankwords2lowercase[:160])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['if', 'the', 'wife', 'and', 'i', 'are', 'fussin', \"'\", ',', 'brother', ',', 'that', \"'s\", 'our', 'right', \"'cause\", 'me', 'and', 'that', 'sweet', 'woman', \"'s\", 'got', 'a', 'license', 'to', 'fight', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', ',', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', \"'\", 'mine', 'oh', ',', 'the', 'woman', 'on', 'our', 'party', 'line', \"'s\", 'the', 'nosiest', 'thing', 'she', 'picks', 'up', 'her', 'receiver', 'when', 'she', 'knows', 'it', \"'s\", 'my', 'ring', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', ',', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', \"'\", 'mine', 'i', 'got', 'a', 'little', 'gal', 'that', 'wears', 'her', 'hair', 'up', 'high', 'the', 'boys', 'all', 'whistle', 'when', 'she', 'walks', 'by', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', 'well', ',', 'if', 'you', 'mind', 'your', 'business', ',', 'then', 'you', 'sure', 'wo', \"n't\", 'be', 'mindin', \"'\", 'mine', 'if', 'i', 'want', 'to']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p2VB4RlJytxN",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# this function takes a string and a list of words as parameters.\n",
        "#   It will return all the words in the list that contain the string as a substring\n",
        "def searchstring (substring, wordlist):\n",
        "    # initialize the result\n",
        "    result = [ ]\n",
        "    #  loop over all the words\n",
        "    for word in wordlist:\n",
        "        # test each word if it contains the substring\n",
        "        if substring in word:\n",
        "            # add it to the result\n",
        "            result.append(word)\n",
        "    return result"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8R1q0NEMyxuM",
        "colab_type": "code",
        "outputId": "2a997f9f-67bd-4a95-b1bc-f4708451da29",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "searchstring('z', hankwords)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['lazarus', \"buzzin'\", 'dozen', 'magazines', 'magazines', 'crazy']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8-HFIRWuy1_x",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "##  Regular Expression to match non-alphabetic characters\n",
        "import re"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PFIpVbe74lC1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# this regular expression pattern matches any word that contains all non-alphabetical\n",
        "#   lower-case characters [^a-z]+\n",
        "# the beginning ^ and ending $ require the match to begin and end on a word boundary \n",
        "pattern = re.compile('^[^a-z]+$')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6Gh9VzRIy4PO",
        "colab_type": "code",
        "outputId": "a96af771-1c98-4340-d229-ab4f271c5e47",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "nonAlphaMatch = pattern.match('**')\n",
        "#  if it matched, print a message\n",
        "if nonAlphaMatch: print ('matched non-alphabetical')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "matched non-alphabetical\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nPAthzWoy6xO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# function that takes a word and returns true if it consists only\n",
        "#   of non-alphabetic characters  (assumes import re)\n",
        "def alpha_filter(w):\n",
        "  # pattern to match word of non-alphabetical characters\n",
        "  pattern = re.compile('^[^a-z]+$')\n",
        "  if (pattern.match(w)):\n",
        "    return True\n",
        "  else:\n",
        "    return False"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sTzzhH8ky9LN",
        "colab_type": "code",
        "outputId": "06aa4d10-0e93-4583-a27c-9f3619d38cc4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# apply the function to hankwords\n",
        "alphahankwords = [w for w in hankwords if not alpha_filter(w)]\n",
        "print(alphahankwords[:100])\n",
        "print(len(alphahankwords))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['if', 'the', 'wife', 'and', 'i', 'are', 'fussin', 'brother', 'that', \"'s\", 'our', 'right', \"'cause\", 'me', 'and', 'that', 'sweet', 'woman', \"'s\", 'got', 'a', 'license', 'to', 'fight', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', 'mine', 'oh', 'the', 'woman', 'on', 'our', 'party', 'line', \"'s\", 'the', 'nosiest', 'thing', 'she', 'picks', 'up', 'her', 'receiver', 'when', 'she', 'knows', 'it', \"'s\", 'my', 'ring', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', 'mine', 'i', 'got', 'a']\n",
            "7473\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4yZo62cmy_ic",
        "colab_type": "code",
        "outputId": "87588585-8543-4aa5-d5e3-eafb5ed13d4b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# get a list of stopwords from nltk\n",
        "nltkstopwords = nltk.corpus.stopwords.words('english')\n",
        "print(len(nltkstopwords))\n",
        "print(nltkstopwords)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "179\n",
            "['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', \"you're\", \"you've\", \"you'll\", \"you'd\", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', 'she', \"she's\", 'her', 'hers', 'herself', 'it', \"it's\", 'its', 'itself', 'they', 'them', 'their', 'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', \"that'll\", 'these', 'those', 'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', 'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', 'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after', 'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further', 'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more', 'most', 'other', 'some', 'such', 'no', 'nor', 'not', 'only', 'own', 'same', 'so', 'than', 'too', 'very', 's', 't', 'can', 'will', 'just', 'don', \"don't\", 'should', \"should've\", 'now', 'd', 'll', 'm', 'o', 're', 've', 'y', 'ain', 'aren', \"aren't\", 'couldn', \"couldn't\", 'didn', \"didn't\", 'doesn', \"doesn't\", 'hadn', \"hadn't\", 'hasn', \"hasn't\", 'haven', \"haven't\", 'isn', \"isn't\", 'ma', 'mightn', \"mightn't\", 'mustn', \"mustn't\", 'needn', \"needn't\", 'shan', \"shan't\", 'shouldn', \"shouldn't\", 'wasn', \"wasn't\", 'weren', \"weren't\", 'won', \"won't\", 'wouldn', \"wouldn't\"]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mUnfy9htzCN4",
        "colab_type": "code",
        "outputId": "3c307b94-96cf-4d08-dac9-0f60625efcae",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# check tokenization in hankwords\n",
        "print(hankwords[:100])\n",
        "print(hankwords[15300:15310])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['if', 'the', 'wife', 'and', 'i', 'are', 'fussin', \"'\", ',', 'brother', ',', 'that', \"'s\", 'our', 'right', \"'cause\", 'me', 'and', 'that', 'sweet', 'woman', \"'s\", 'got', 'a', 'license', 'to', 'fight', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', ',', 'then', 'you', 'wo', \"n't\", 'be', 'mindin', \"'\", 'mine', 'oh', ',', 'the', 'woman', 'on', 'our', 'party', 'line', \"'s\", 'the', 'nosiest', 'thing', 'she', 'picks', 'up', 'her', 'receiver', 'when', 'she', 'knows', 'it', \"'s\", 'my', 'ring', 'why', 'do', \"n't\", 'you', 'mind', 'your', 'own', 'business', '?', 'mind', 'your', 'own', 'business', \"'cause\", 'if', 'you', 'mind', 'your', 'business', ',', 'then']\n",
            "[]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f8NdlU2o0CiS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "morestopwords = ['hank','williams','could','would','might','must','need','sha','wo','y',\"'s\",\"'d\",\"'ll\",\"'t\",\"'m\",\"'re\",\"'ve\", \"n't\", 'repeat','chorus','verse']"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NXyhpiObzE_q",
        "colab_type": "code",
        "outputId": "88e4b0b3-d08a-4fff-cc03-13b4b1bad4bc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "stopwords = nltkstopwords + morestopwords\n",
        "print(len(stopwords))\n",
        "print(stopwords)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "200\n",
            "['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', \"you're\", \"you've\", \"you'll\", \"you'd\", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', 'she', \"she's\", 'her', 'hers', 'herself', 'it', \"it's\", 'its', 'itself', 'they', 'them', 'their', 'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', \"that'll\", 'these', 'those', 'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', 'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', 'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after', 'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further', 'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more', 'most', 'other', 'some', 'such', 'no', 'nor', 'not', 'only', 'own', 'same', 'so', 'than', 'too', 'very', 's', 't', 'can', 'will', 'just', 'don', \"don't\", 'should', \"should've\", 'now', 'd', 'll', 'm', 'o', 're', 've', 'y', 'ain', 'aren', \"aren't\", 'couldn', \"couldn't\", 'didn', \"didn't\", 'doesn', \"doesn't\", 'hadn', \"hadn't\", 'hasn', \"hasn't\", 'haven', \"haven't\", 'isn', \"isn't\", 'ma', 'mightn', \"mightn't\", 'mustn', \"mustn't\", 'needn', \"needn't\", 'shan', \"shan't\", 'shouldn', \"shouldn't\", 'wasn', \"wasn't\", 'weren', \"weren't\", 'won', \"won't\", 'wouldn', \"wouldn't\", 'hank', 'williams', 'could', 'would', 'might', 'must', 'need', 'sha', 'wo', 'y', \"'s\", \"'d\", \"'ll\", \"'t\", \"'m\", \"'re\", \"'ve\", \"n't\", 'repeat', 'chorus', 'verse']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b99MiqcgzHWH",
        "colab_type": "code",
        "outputId": "e8d42c49-3dd2-458f-aab6-e88373e75a1a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "stoppedhankwords = [w for w in alphahankwords if not w in stopwords]\n",
        "print(len(stoppedhankwords))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "3435\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6od5xTQ2zJiK",
        "colab_type": "code",
        "outputId": "2086fbe6-1143-4037-a337-289a31001933",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        }
      },
      "source": [
        "# use this list for a better frequency distribution\n",
        "hankdist = FreqDist(stoppedhankwords)\n",
        "hankitems = hankdist.most_common(50)\n",
        "for item in hankitems:\n",
        "  print(item)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "('got', 56)\n",
            "('lord', 48)\n",
            "('love', 44)\n",
            "('heart', 36)\n",
            "('oh', 30)\n",
            "('like', 30)\n",
            "('new', 26)\n",
            "('old', 25)\n",
            "('life', 25)\n",
            "('never', 25)\n",
            "('move', 23)\n",
            "('honky', 22)\n",
            "('mind', 21)\n",
            "('say', 21)\n",
            "('day', 21)\n",
            "('know', 21)\n",
            "('water', 20)\n",
            "('well', 19)\n",
            "('praise', 19)\n",
            "('take', 19)\n",
            "('free', 18)\n",
            "('go', 17)\n",
            "('blues', 17)\n",
            "('ca', 17)\n",
            "('light', 17)\n",
            "('away', 17)\n",
            "('business', 16)\n",
            "('man', 16)\n",
            "('see', 16)\n",
            "('bucket', 16)\n",
            "('soul', 16)\n",
            "('dog', 16)\n",
            "('let', 15)\n",
            "('lost', 15)\n",
            "('yea', 15)\n",
            "('hole', 15)\n",
            "('home', 14)\n",
            "('made', 14)\n",
            "('long', 14)\n",
            "('cry', 14)\n",
            "('cold', 14)\n",
            "('tonk', 13)\n",
            "('god', 13)\n",
            "('gone', 13)\n",
            "('come', 13)\n",
            "('used', 13)\n",
            "('lonesome', 13)\n",
            "('one', 13)\n",
            "('kaw-liga', 13)\n",
            "('mine', 12)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5ortWpq70RNx",
        "colab_type": "code",
        "outputId": "d00864d3-f245-4496-e17b-e91220feb4b5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        }
      },
      "source": [
        "hankdist.plot(30,cumulative=False)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEsCAYAAADO7LQXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "06XQvRMHzLz6",
        "colab_type": "code",
        "outputId": "c8ffbdd3-fc97-4a09-9efd-2a6cadf3c74c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# Bigrams and Bigram frequency distribution\n",
        "hankbigrams = list(nltk.bigrams(hankwords))\n",
        "print(hankwords[:21])\n",
        "print(hankbigrams[:20])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['if', 'the', 'wife', 'and', 'i', 'are', 'fussin', \"'\", ',', 'brother', ',', 'that', \"'s\", 'our', 'right', \"'cause\", 'me', 'and', 'that', 'sweet', 'woman']\n",
            "[('if', 'the'), ('the', 'wife'), ('wife', 'and'), ('and', 'i'), ('i', 'are'), ('are', 'fussin'), ('fussin', \"'\"), (\"'\", ','), (',', 'brother'), ('brother', ','), (',', 'that'), ('that', \"'s\"), (\"'s\", 'our'), ('our', 'right'), ('right', \"'cause\"), (\"'cause\", 'me'), ('me', 'and'), ('and', 'that'), ('that', 'sweet'), ('sweet', 'woman')]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0iBGN8DszNwm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# setup for bigrams and bigram measures\n",
        "from nltk.collocations import *\n",
        "bigram_measures = nltk.collocations.BigramAssocMeasures()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v-gl_-AozP06",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# create the bigram finder and score the bigrams by frequency\n",
        "# BIGRAMS MUST BE APPLIED TO COMPLETE WORD LIST WITHOUT REMOVED STOP WORDS -- LOWER CASE OK\n",
        "finder = BigramCollocationFinder.from_words(hankwords)\n",
        "scored = finder.score_ngrams(bigram_measures.raw_freq)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "29pIS-JjzSIS",
        "colab_type": "code",
        "outputId": "e55ad452-a8fa-48c9-e06a-2255ce30363b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 69
        }
      },
      "source": [
        "# scored is a list of bigram pairs with their score\n",
        "print(type(scored))\n",
        "first = scored[0]\n",
        "print(type(first))\n",
        "print(first)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'list'>\n",
            "<class 'tuple'>\n",
            "(('i', \"'m\"), 0.006459627329192546)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iZe3zhCNzUYe",
        "colab_type": "code",
        "outputId": "7faf20d3-e9b6-40ec-e2fc-b9e3a52149fa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        }
      },
      "source": [
        "# scores are sorted in decreasing frequency\n",
        "for bscore in scored[:50]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(('i', \"'m\"), 0.006459627329192546)\n",
            "((\"''\", \"''\"), 0.005714285714285714)\n",
            "(('i', \"'ll\"), 0.0055900621118012426)\n",
            "((',', 'i'), 0.004720496894409938)\n",
            "(('on', 'the'), 0.003975155279503106)\n",
            "((\"'ll\", 'have'), 0.0036024844720496892)\n",
            "(('it', 'on'), 0.0036024844720496892)\n",
            "(('do', \"n't\"), 0.0034782608695652175)\n",
            "(('on', 'over'), 0.0034782608695652175)\n",
            "(('in', 'the'), 0.0033540372670807454)\n",
            "(('have', 'a'), 0.0029813664596273293)\n",
            "(('a', 'new'), 0.002857142857142857)\n",
            "(('got', 'a'), 0.002857142857142857)\n",
            "(('lord', ','), 0.002857142857142857)\n",
            "(('we', \"'ll\"), 0.002732919254658385)\n",
            "(('i', 'got'), 0.002484472049689441)\n",
            "((\"'s\", 'got'), 0.002360248447204969)\n",
            "(('and', 'i'), 0.0022360248447204968)\n",
            "(('ca', \"n't\"), 0.002111801242236025)\n",
            "(('the', 'lord'), 0.002111801242236025)\n",
            "(('in', 'it'), 0.001987577639751553)\n",
            "(('move', 'it'), 0.001987577639751553)\n",
            "(('my', 'heart'), 0.001987577639751553)\n",
            "((\"n't\", 'you'), 0.001987577639751553)\n",
            "(('you', \"'ll\"), 0.001987577639751553)\n",
            "(('!', 'my'), 0.0018633540372670807)\n",
            "((',', 'he'), 0.0018633540372670807)\n",
            "(('a', 'hole'), 0.0018633540372670807)\n",
            "(('bucket', \"'s\"), 0.0018633540372670807)\n",
            "(('hole', 'in'), 0.0018633540372670807)\n",
            "(('mind', 'your'), 0.0018633540372670807)\n",
            "(('my', 'bucket'), 0.0018633540372670807)\n",
            "(('no', 'more'), 0.0018633540372670807)\n",
            "(('praise', 'the'), 0.0018633540372670807)\n",
            "(('the', 'light'), 0.0018633540372670807)\n",
            "(('yea', '!'), 0.0018633540372670807)\n",
            "((\"'ll\", 'be'), 0.0017391304347826088)\n",
            "(('(', 'move'), 0.0017391304347826088)\n",
            "(('all', 'the'), 0.0017391304347826088)\n",
            "(('oh', ','), 0.0017391304347826088)\n",
            "(('over', '('), 0.0017391304347826088)\n",
            "(('over', ')'), 0.0017391304347826088)\n",
            "(('that', 'you'), 0.0017391304347826088)\n",
            "(('you', \"'re\"), 0.0017391304347826088)\n",
            "(('it', \"'s\"), 0.0016149068322981366)\n",
            "(('lord', 'i'), 0.0016149068322981366)\n",
            "(('used', 'to'), 0.0016149068322981366)\n",
            "(('why', 'do'), 0.0016149068322981366)\n",
            "(('i', 'saw'), 0.0014906832298136647)\n",
            "(('if', 'you'), 0.0014906832298136647)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DcEFunUzzWf4",
        "colab_type": "code",
        "outputId": "1dd79b8c-a19c-4587-da80-f6c3196f6443",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 537
        }
      },
      "source": [
        "# apply a filter to remove non-alphabetical tokens from the hank bigram finder\n",
        "finder.apply_word_filter(alpha_filter)\n",
        "scored = finder.score_ngrams(bigram_measures.raw_freq)\n",
        "for bscore in scored[:30]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(('i', \"'m\"), 0.006459627329192546)\n",
            "(('i', \"'ll\"), 0.0055900621118012426)\n",
            "(('on', 'the'), 0.003975155279503106)\n",
            "((\"'ll\", 'have'), 0.0036024844720496892)\n",
            "(('it', 'on'), 0.0036024844720496892)\n",
            "(('do', \"n't\"), 0.0034782608695652175)\n",
            "(('on', 'over'), 0.0034782608695652175)\n",
            "(('in', 'the'), 0.0033540372670807454)\n",
            "(('have', 'a'), 0.0029813664596273293)\n",
            "(('a', 'new'), 0.002857142857142857)\n",
            "(('got', 'a'), 0.002857142857142857)\n",
            "(('we', \"'ll\"), 0.002732919254658385)\n",
            "(('i', 'got'), 0.002484472049689441)\n",
            "((\"'s\", 'got'), 0.002360248447204969)\n",
            "(('and', 'i'), 0.0022360248447204968)\n",
            "(('ca', \"n't\"), 0.002111801242236025)\n",
            "(('the', 'lord'), 0.002111801242236025)\n",
            "(('in', 'it'), 0.001987577639751553)\n",
            "(('move', 'it'), 0.001987577639751553)\n",
            "(('my', 'heart'), 0.001987577639751553)\n",
            "((\"n't\", 'you'), 0.001987577639751553)\n",
            "(('you', \"'ll\"), 0.001987577639751553)\n",
            "(('a', 'hole'), 0.0018633540372670807)\n",
            "(('bucket', \"'s\"), 0.0018633540372670807)\n",
            "(('hole', 'in'), 0.0018633540372670807)\n",
            "(('mind', 'your'), 0.0018633540372670807)\n",
            "(('my', 'bucket'), 0.0018633540372670807)\n",
            "(('no', 'more'), 0.0018633540372670807)\n",
            "(('praise', 'the'), 0.0018633540372670807)\n",
            "(('the', 'light'), 0.0018633540372670807)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OivXykKgzYrS",
        "colab_type": "code",
        "outputId": "f4a7e8ca-1d07-4552-bef4-d90e572cef66",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        }
      },
      "source": [
        "# apply a filter to remove stop words\n",
        "# lambda is a function definition 'on the fly'\n",
        "finder.apply_word_filter(lambda w: w in stopwords)\n",
        "scored = finder.score_ngrams(bigram_measures.raw_freq)\n",
        "for bscore in scored[:50]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(('tonk', 'blues'), 0.0014906832298136647)\n",
            "(('body', 'praise'), 0.0011180124223602484)\n",
            "(('gon', 'na'), 0.0011180124223602484)\n",
            "(('honky', 'tonk'), 0.0011180124223602484)\n",
            "(('honky', 'tonkin'), 0.0011180124223602484)\n",
            "(('new', 'body'), 0.0011180124223602484)\n",
            "(('new', 'life'), 0.0011180124223602484)\n",
            "(('clear', 'water'), 0.0009937888198757764)\n",
            "(('big', 'fun'), 0.0008695652173913044)\n",
            "(('blues', 'well'), 0.0008695652173913044)\n",
            "(('dog', 'cause'), 0.0008695652173913044)\n",
            "(('precious', 'lord'), 0.0008695652173913044)\n",
            "(('country', 'church'), 0.0007453416149068323)\n",
            "(('go', 'honky'), 0.0007453416149068323)\n",
            "(('oh', 'lord'), 0.0007453416149068323)\n",
            "(('old', 'country'), 0.0007453416149068323)\n",
            "(('poor', 'old'), 0.0007453416149068323)\n",
            "(('fly', 'away'), 0.0006211180124223603)\n",
            "(('got', 'ta'), 0.0006211180124223603)\n",
            "(('never', 'let'), 0.0006211180124223603)\n",
            "(('old', 'kaw-liga'), 0.0006211180124223603)\n",
            "(('poor', 'soul'), 0.0006211180124223603)\n",
            "(('wedding', 'bells'), 0.0006211180124223603)\n",
            "(('bayou', 'jambalaya'), 0.0004968944099378882)\n",
            "(('beer', 'well'), 0.0004968944099378882)\n",
            "(('blues', 'yeah'), 0.0004968944099378882)\n",
            "(('cold', 'heart'), 0.0004968944099378882)\n",
            "(('crawfish', 'pie'), 0.0004968944099378882)\n",
            "(('doubtful', 'mind'), 0.0004968944099378882)\n",
            "(('god', 'dips'), 0.0004968944099378882)\n",
            "(('hand', 'precious'), 0.0004968944099378882)\n",
            "(('ho-on-ky', 'tonk'), 0.0004968944099378882)\n",
            "(('honey', 'baby'), 0.0004968944099378882)\n",
            "(('lonesome', 'blue'), 0.0004968944099378882)\n",
            "(('long', 'gone'), 0.0004968944099378882)\n",
            "(('lost', 'highway'), 0.0004968944099378882)\n",
            "(('said', 'goodbye'), 0.0004968944099378882)\n",
            "(('saviour', 'call'), 0.0004968944099378882)\n",
            "(('sight', 'praise'), 0.0004968944099378882)\n",
            "(('well', 'lord'), 0.0004968944099378882)\n",
            "((\"'cause\", 'tonight'), 0.00037267080745341616)\n",
            "(('amio', 'pick'), 0.00037267080745341616)\n",
            "(('bad', 'news'), 0.00037267080745341616)\n",
            "(('carry', 'bad'), 0.00037267080745341616)\n",
            "(('cher', 'amio'), 0.00037267080745341616)\n",
            "(('die', 'like'), 0.00037267080745341616)\n",
            "(('dying', 'breath'), 0.00037267080745341616)\n",
            "(('fall', 'praise'), 0.00037267080745341616)\n",
            "(('filled', 'fruit'), 0.00037267080745341616)\n",
            "(('filé', 'gumbo'), 0.00037267080745341616)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6gooeeoczbTE",
        "colab_type": "code",
        "outputId": "6e7c4b1b-e869-4684-c8aa-22fdbb801d4c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        }
      },
      "source": [
        "# apply a filter (on a new finder) to remove low frequency words\n",
        "finder2 = BigramCollocationFinder.from_words(hankwords)\n",
        "finder2.apply_freq_filter(2)\n",
        "scored = finder2.score_ngrams(bigram_measures.raw_freq)\n",
        "for bscore in scored[:20]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(('i', \"'m\"), 0.006459627329192546)\n",
            "((\"''\", \"''\"), 0.005714285714285714)\n",
            "(('i', \"'ll\"), 0.0055900621118012426)\n",
            "((',', 'i'), 0.004720496894409938)\n",
            "(('on', 'the'), 0.003975155279503106)\n",
            "((\"'ll\", 'have'), 0.0036024844720496892)\n",
            "(('it', 'on'), 0.0036024844720496892)\n",
            "(('do', \"n't\"), 0.0034782608695652175)\n",
            "(('on', 'over'), 0.0034782608695652175)\n",
            "(('in', 'the'), 0.0033540372670807454)\n",
            "(('have', 'a'), 0.0029813664596273293)\n",
            "(('a', 'new'), 0.002857142857142857)\n",
            "(('got', 'a'), 0.002857142857142857)\n",
            "(('lord', ','), 0.002857142857142857)\n",
            "(('we', \"'ll\"), 0.002732919254658385)\n",
            "(('i', 'got'), 0.002484472049689441)\n",
            "((\"'s\", 'got'), 0.002360248447204969)\n",
            "(('and', 'i'), 0.0022360248447204968)\n",
            "(('ca', \"n't\"), 0.002111801242236025)\n",
            "(('the', 'lord'), 0.002111801242236025)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z0sf5CRfzdnm",
        "colab_type": "code",
        "outputId": "0cca9ae7-b008-4e5c-b63d-8f6d7d5a1643",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        }
      },
      "source": [
        "# apply a filter on both words of the ngram\n",
        "finder2.apply_ngram_filter(lambda w1, w2: len(w1) < 2)\n",
        "scored = finder2.score_ngrams(bigram_measures.raw_freq)\n",
        "for bscore in scored[:20]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "((\"''\", \"''\"), 0.005714285714285714)\n",
            "(('on', 'the'), 0.003975155279503106)\n",
            "((\"'ll\", 'have'), 0.0036024844720496892)\n",
            "(('it', 'on'), 0.0036024844720496892)\n",
            "(('do', \"n't\"), 0.0034782608695652175)\n",
            "(('on', 'over'), 0.0034782608695652175)\n",
            "(('in', 'the'), 0.0033540372670807454)\n",
            "(('have', 'a'), 0.0029813664596273293)\n",
            "(('got', 'a'), 0.002857142857142857)\n",
            "(('lord', ','), 0.002857142857142857)\n",
            "(('we', \"'ll\"), 0.002732919254658385)\n",
            "((\"'s\", 'got'), 0.002360248447204969)\n",
            "(('and', 'i'), 0.0022360248447204968)\n",
            "(('ca', \"n't\"), 0.002111801242236025)\n",
            "(('the', 'lord'), 0.002111801242236025)\n",
            "(('in', 'it'), 0.001987577639751553)\n",
            "(('move', 'it'), 0.001987577639751553)\n",
            "(('my', 'heart'), 0.001987577639751553)\n",
            "((\"n't\", 'you'), 0.001987577639751553)\n",
            "(('you', \"'ll\"), 0.001987577639751553)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vSihyFp_zfvA",
        "colab_type": "code",
        "outputId": "4ba19995-34d9-43eb-c195-5839e27b8fe6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        }
      },
      "source": [
        "### pointwise mutual information\n",
        "### THIS IS THE WRONG WAY TO DO THIS -- ONLY SCORES THINGS THAT OCCUR TOGETHER SUCH AS COMMON WORDS\n",
        "### APPLY A FREQUENCY FILTER OF ABOUT 5 AS SHOWN BELOW\n",
        "finder3 = BigramCollocationFinder.from_words(hankwords)\n",
        "scored = finder3.score_ngrams(bigram_measures.pmi)\n",
        "for bscore in scored[:50]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "((\"allowin'\", 'tomorrow'), 12.974773067889343)\n",
            "(('alright', 'no-how'), 12.974773067889343)\n",
            "(('awful', 'shape'), 12.974773067889343)\n",
            "(('barren', 'waste'), 12.974773067889343)\n",
            "(('blossoms', 'unfold'), 12.974773067889343)\n",
            "(('bursting', 'saints'), 12.974773067889343)\n",
            "((\"buzzin'\", 'kinfolk'), 12.974773067889343)\n",
            "(('car', 'speeded'), 12.974773067889343)\n",
            "(('children', 'laughing'), 12.974773067889343)\n",
            "(('coal', 'black'), 12.974773067889343)\n",
            "(('crushed', 'broken'), 12.974773067889343)\n",
            "(('distant', 'uncle'), 12.974773067889343)\n",
            "(('divine', 'fills'), 12.974773067889343)\n",
            "(('doggish', 'ways'), 12.974773067889343)\n",
            "(('dollar', 'bill'), 12.974773067889343)\n",
            "(('dozen', 'dressed'), 12.974773067889343)\n",
            "(('evil', 'scheme'), 12.974773067889343)\n",
            "(('feller', 'dress'), 12.974773067889343)\n",
            "(('future', 'looks'), 12.974773067889343)\n",
            "(('gabriel', 'blows'), 12.974773067889343)\n",
            "(('goes', 'crawling'), 12.974773067889343)\n",
            "(('grow', 'dim'), 12.974773067889343)\n",
            "(('grows', 'drear'), 12.974773067889343)\n",
            "(('hog', 'wild'), 12.974773067889343)\n",
            "(('hollered', 'whoa'), 12.974773067889343)\n",
            "(('laid', 'beneath'), 12.974773067889343)\n",
            "(('lawyer', 'proved'), 12.974773067889343)\n",
            "(('many', 'faults'), 12.974773067889343)\n",
            "(('might', 'sa-ay'), 12.974773067889343)\n",
            "(('mighty', 'small'), 12.974773067889343)\n",
            "(('monkey', 'wrench'), 12.974773067889343)\n",
            "(('mornin', 'wishin'), 12.974773067889343)\n",
            "(('nearly', 'twenty-two'), 12.974773067889343)\n",
            "(('open', 'ro-oad'), 12.974773067889343)\n",
            "(('organ', 'plays'), 12.974773067889343)\n",
            "(('probably', 'drown'), 12.974773067889343)\n",
            "(('rails', \"'neath\"), 12.974773067889343)\n",
            "(('remember', 'pup'), 12.974773067889343)\n",
            "(('ro-oad', 'starts'), 12.974773067889343)\n",
            "(('rolling', 'stone'), 12.974773067889343)\n",
            "(('running', 'board'), 12.974773067889343)\n",
            "(('rural', 'route'), 12.974773067889343)\n",
            "(('sign', 'because'), 12.974773067889343)\n",
            "(('slowly', \"stealin'\"), 12.974773067889343)\n",
            "(('soda', 'pop'), 12.974773067889343)\n",
            "(('spin', 'to-day'), 12.974773067889343)\n",
            "(('summer', 'grew'), 12.974773067889343)\n",
            "(('thibodaux', 'fontaineaux'), 12.974773067889343)\n",
            "(('throats', 'burned'), 12.974773067889343)\n",
            "((\"tookin'\", 'hows'), 12.974773067889343)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Jz_RNEgWzhym",
        "colab_type": "code",
        "outputId": "4b915b84-3190-44db-e2f2-396ab15f76e0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 416
        }
      },
      "source": [
        "# to get good results, must first apply frequency filter\n",
        "finder.apply_freq_filter(5)\n",
        "scored = finder.score_ngrams(bigram_measures.pmi)\n",
        "for bscore in scored[:50]:\n",
        "    print (bscore)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(('wedding', 'bells'), 10.389810567168187)\n",
            "(('country', 'church'), 9.974773067889341)\n",
            "(('gon', 'na'), 9.65284497300198)\n",
            "(('big', 'fun'), 9.290274893617273)\n",
            "(('tonk', 'blues'), 8.771833009219067)\n",
            "(('clear', 'water'), 8.652844973001981)\n",
            "(('fly', 'away'), 8.62427582080521)\n",
            "(('body', 'praise'), 8.574842461000706)\n",
            "(('honky', 'tonkin'), 8.515341449252045)\n",
            "(('dog', 'cause'), 8.32269637130965)\n",
            "(('new', 'body'), 8.122330256303199)\n",
            "(('honky', 'tonk'), 7.984826732553264)\n",
            "(('old', 'country'), 7.915879378835775)\n",
            "(('poor', 'soul'), 7.837269544139407)\n",
            "(('poor', 'old'), 7.4564477601984755)\n",
            "(('blues', 'well'), 7.446737635253022)\n",
            "(('got', 'ta'), 7.167418145831739)\n",
            "(('precious', 'lord'), 7.027240487783477)\n",
            "(('go', 'honky'), 7.012841108722862)\n",
            "(('old', 'kaw-liga'), 6.952405254860889)\n",
            "(('new', 'life'), 6.800402161415835)\n",
            "(('never', 'let'), 6.745954377393462)\n",
            "(('oh', 'lord'), 5.067882472280823)\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}