{"nbformat": 3, "worksheets": [{"cells": [{"cell_type": "code", "language": "python", "input": ["import numpy as np\n", "import numpy.linalg as la\n", "import matplotlib.pyplot as pt"], "collapsed": false, "prompt_number": 9, "metadata": {}, "outputs": []}, {"cell_type": "code", "language": "python", "input": ["with open(\"nixon.txt\") as textf:\n", "    nixon = textf.read()\n", "with open(\"clinton.txt\") as textf:\n", "    clinton = textf.read()\n", "with open(\"rail-transport.txt\") as textf:\n", "    rail = textf.read()"], "collapsed": false, "prompt_number": 10, "metadata": {}, "outputs": []}, {"cell_type": "code", "language": "python", "input": ["print(nixon[:500])"], "collapsed": false, "prompt_number": 11, "metadata": {}, "outputs": [{"stream": "stdout", "text": ["\n", "Richard Nixon\n", "From Wikipedia, the free encyclopedia\n", "Richard Nixon\n", "Richard M. Nixon, ca. 1935 - 1982 - NARA - 530679.jpg\n", "37th President of the United States\n", "In office\n", "January 20, 1969 \u2013 August 9, 1974\n", "Vice President  \n", "\n", "    Spiro Agnew (1969\u20131973)\n", "    None (Oct\u2013Dec 1973)\n", "    Gerald Ford (1973\u20131974)\n", "\n", "Preceded by     Lyndon Johnson\n", "Succeeded by    Gerald Ford\n", "36th Vice President of the United States\n", "In office\n", "January 20, 1953 \u2013 January 20, 1961\n", "President       Dwight Eisenhower\n", "Preceded b\n"], "output_type": "stream"}]}, {"source": ["How would you turn this into a vector?"], "metadata": {}, "cell_type": "markdown"}, {"cell_type": "code", "language": "python", "input": ["nixon_vec = [ord(c) for c in nixon]"], "collapsed": false, "prompt_number": 12, "metadata": {}, "outputs": []}, {"cell_type": "code", "language": "python", "input": ["print nixon_vec[:300]"], "collapsed": false, "prompt_number": 13, "metadata": {}, "outputs": [{"stream": "stdout", "text": ["[10, 82, 105, 99, 104, 97, 114, 100, 32, 78, 105, 120, 111, 110, 10, 70, 114, 111, 109, 32, 87, 105, 107, 105, 112, 101, 100, 105, 97, 44, 32, 116, 104, 101, 32, 102, 114, 101, 101, 32, 101, 110, 99, 121, 99, 108, 111, 112, 101, 100, 105, 97, 10, 82, 105, 99, 104, 97, 114, 100, 32, 78, 105, 120, 111, 110, 10, 82, 105, 99, 104, 97, 114, 100, 32, 77, 46, 32, 78, 105, 120, 111, 110, 44, 32, 99, 97, 46, 32, 49, 57, 51, 53, 32, 45, 32, 49, 57, 56, 50, 32, 45, 32, 78, 65, 82, 65, 32, 45, 32, 53, 51, 48, 54, 55, 57, 46, 106, 112, 103, 10, 51, 55, 116, 104, 32, 80, 114, 101, 115, 105, 100, 101, 110, 116, 32, 111, 102, 32, 116, 104, 101, 32, 85, 110, 105, 116, 101, 100, 32, 83, 116, 97, 116, 101, 115, 10, 73, 110, 32, 111, 102, 102, 105, 99, 101, 10, 74, 97, 110, 117, 97, 114, 121, 32, 50, 48, 44, 32, 49, 57, 54, 57, 32, 226, 128, 147, 32, 65, 117, 103, 117, 115, 116, 32, 57, 44, 32, 49, 57, 55, 52, 10, 86, 105, 99, 101, 32, 80, 114, 101, 115, 105, 100, 101, 110, 116, 32, 32, 10, 10, 32, 32, 32, 32, 83, 112, 105, 114, 111, 32, 65, 103, 110, 101, 119, 32, 40, 49, 57, 54, 57, 226, 128, 147, 49, 57, 55, 51, 41, 10, 32, 32, 32, 32, 78, 111, 110, 101, 32, 40, 79, 99, 116, 226, 128, 147, 68, 101, 99, 32, 49, 57, 55, 51, 41, 10, 32, 32, 32, 32, 71, 101, 114, 97, 108, 100, 32, 70, 111, 114, 100, 32, 40, 49, 57, 55, 51, 226, 128]\n"], "output_type": "stream"}]}, {"source": ["Let's try another way."], "metadata": {}, "cell_type": "markdown"}, {"cell_type": "code", "language": "python", "input": ["def split_into_words(s):\n", "    import re\n", "    return re.split(\"[., \\n\\t]+\", s)"], "collapsed": false, "prompt_number": 14, "metadata": {}, "outputs": []}, {"cell_type": "code", "language": "python", "input": ["print split_into_words(nixon)[:300]"], "collapsed": false, "prompt_number": 17, "metadata": {}, "outputs": [{"stream": "stdout", "text": ["['', 'Richard', 'Nixon', 'From', 'Wikipedia', 'the', 'free', 'encyclopedia', 'Richard', 'Nixon', 'Richard', 'M', 'Nixon', 'ca', '1935', '-', '1982', '-', 'NARA', '-', '530679', 'jpg', '37th', 'President', 'of', 'the', 'United', 'States', 'In', 'office', 'January', '20', '1969', '\\xe2\\x80\\x93', 'August', '9', '1974', 'Vice', 'President', 'Spiro', 'Agnew', '(1969\\xe2\\x80\\x931973)', 'None', '(Oct\\xe2\\x80\\x93Dec', '1973)', 'Gerald', 'Ford', '(1973\\xe2\\x80\\x931974)', 'Preceded', 'by', 'Lyndon', 'Johnson', 'Succeeded', 'by', 'Gerald', 'Ford', '36th', 'Vice', 'President', 'of', 'the', 'United', 'States', 'In', 'office', 'January', '20', '1953', '\\xe2\\x80\\x93', 'January', '20', '1961', 'President', 'Dwight', 'Eisenhower', 'Preceded', 'by', 'Alben', 'Barkley', 'Succeeded', 'by', 'Lyndon', 'Johnson', 'United', 'States', 'Senator', 'from', 'California', 'In', 'office', 'December', '4', '1950', '\\xe2\\x80\\x93', 'January', '1', '1953', 'Preceded', 'by', 'Sheridan', 'Downey', 'Succeeded', 'by', 'Thomas', 'Kuchel', 'Member', 'of', 'the', 'U', 'S', 'House', 'of', 'Representatives', 'from', \"California's\", '12th', 'district', 'In', 'office', 'January', '3', '1947', '\\xe2\\x80\\x93', 'December', '1', '1950', 'Preceded', 'by', 'Jerry', 'Voorhis', 'Succeeded', 'by', 'Patrick', 'Hillings', 'Personal', 'details', 'Born', 'Richard', 'Milhous', 'Nixon', 'January', '9', '1913', 'Yorba', 'Linda', 'California', 'U', 'S', 'Died', 'April', '22', '1994', '(aged', '81)', 'New', 'York', 'City', 'New', 'York', 'U', 'S', 'Resting', 'place', 'Richard', 'Nixon', 'Presidential', 'Library', 'and', 'Museum', 'Yorba', 'Linda', 'California', 'U', 'S', 'Political', 'party', 'Republican', 'Spouse(s)', 'Pat', 'Ryan', '(1940\\xe2\\x80\\x931993;', 'her', 'death)', 'Children', 'Tricia', 'Nixon', 'Julie', 'Nixon', 'Alma', 'mater', 'Whittier', 'College', 'Duke', 'University', 'School', 'of', 'Law', 'Profession', 'Lawyer', 'Religion', 'Quaker', 'Nickname', 'Dick', 'Signature', 'Cursive', 'signature', 'in', 'ink', 'Military', 'service', 'Allegiance', 'United', 'States', 'of', 'America', 'Service/branch', 'United', 'States', 'Navy', 'Years', 'of', 'service', '1942\\xe2\\x80\\x931946', 'Rank', 'US-O4', 'insignia', 'svg', 'Lieutenant', 'commander', 'Battles/wars', 'World', 'War', 'II', '\\xe2\\x80\\xa2', 'Pacific', 'War', 'Awards', 'American', 'Campaign', 'Medal', 'Asiatic-Pacific', 'Campaign', 'Medal', '(2', 'service', 'stars)', 'World', 'War', 'II', 'Victory', 'Medal', 'Richard', 'Milhous', 'Nixon', '(January', '9', '1913', '\\xe2\\x80\\x93', 'April', '22', '1994)', 'was', 'the', '37th', 'President', 'of', 'the', 'United', 'States', 'serving', 'from', '1969', 'to', '1974', 'when', 'he', 'became', 'the', 'only', 'president', 'to', 'resign', 'the', 'office', 'Nixon', 'had', 'previously', 'served', 'as', 'a', 'Republican', 'U', 'S', 'Representative', 'and', 'Senator', 'from', 'California', 'and', 'as']\n"], "output_type": "stream"}]}, {"cell_type": "code", "language": "python", "input": ["\n", "word_numbers = {}\n", "\n", "def learn(s):\n", "    for word in split_into_words(s):\n", "        if word not in word_numbers:\n", "            word_numbers[word] = len(word_numbers)\n", "            \n", "def string_to_vec(s):\n", "    result = np.zeros(len(word_numbers))\n", "    for word in split_into_words(s):\n", "        result[word_numbers[word]] += 1\n", "    return result"], "collapsed": false, "prompt_number": 18, "metadata": {}, "outputs": []}, {"cell_type": "code", "language": "python", "input": ["\n", "learn(nixon)\n", "learn(clinton)\n", "learn(rail)\n", "\n", "nixon_v = string_to_vec(nixon)\n", "clinton_v = string_to_vec(clinton)\n", "rail_v = string_to_vec(rail)"], "collapsed": false, "prompt_number": 19, "metadata": {}, "outputs": []}, {"source": ["Plotting could give some insight on this data."], "metadata": {}, "cell_type": "markdown"}, {"cell_type": "code", "language": "python", "input": ["pt.plot(nixon_v)"], "collapsed": false, "prompt_number": 20, "metadata": {}, "outputs": [{"text": ["[<matplotlib.lines.Line2D at 0x7f2970fb1510>]"], "prompt_number": 20, "output_type": "pyout", "metadata": {}}, {"text": ["<matplotlib.figure.Figure at 0x7f2971148c10>"], "png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEACAYAAAC+gnFaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJElqVP8PGuI1sWAEWU8AAAAASUVORK5CYII=\n", "output_type": "display_data", "metadata": {}}]}, {"cell_type": "code", "language": "python", "input": ["pt.plot(clinton_v)"], "collapsed": false, "prompt_number": 21, "metadata": {}, "outputs": [{"text": ["[<matplotlib.lines.Line2D at 0x7f2970deaf90>]"], "prompt_number": 21, "output_type": "pyout", "metadata": {}}, {"text": ["<matplotlib.figure.Figure at 0x7f2970f1c950>"], "png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEACAYAAAC+gnFaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkiRJkiRJkiRJkiRJkiRJkiRJkqS0+v8FoKfNohU+eAAAAABJRU5ErkJggg==\n", "output_type": "display_data", "metadata": {}}]}, {"cell_type": "code", "language": "python", "input": ["pt.plot(rail_v)"], "collapsed": false, "prompt_number": 22, "metadata": {}, "outputs": [{"text": ["[<matplotlib.lines.Line2D at 0x7f2970d2bbd0>]"], "prompt_number": 22, "output_type": "pyout", "metadata": {}}, {"text": ["<matplotlib.figure.Figure at 0x7f2970d84690>"], "png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEACAYAAAC+gnFaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n", "output_type": "display_data", "metadata": {}}]}, {"source": ["Now we need a measure of similarity."], "metadata": {}, "cell_type": "markdown"}, {"cell_type": "code", "language": "python", "input": ["def angle_cos(x, y):\n", "    return np.dot(x, y)/(la.norm(x)*la.norm(y))"], "collapsed": false, "prompt_number": 23, "metadata": {}, "outputs": []}, {"cell_type": "code", "language": "python", "input": ["print angle_cos(nixon_v, clinton_v)"], "collapsed": false, "prompt_number": 24, "metadata": {}, "outputs": [{"stream": "stdout", "text": ["0.858089212872\n"], "output_type": "stream"}]}, {"cell_type": "code", "language": "python", "input": ["print angle_cos(clinton_v, rail_v)"], "collapsed": false, "prompt_number": 27, "metadata": {}, "outputs": [{"stream": "stdout", "text": ["0.835685969275\n"], "output_type": "stream"}]}, {"cell_type": "code", "language": "python", "input": ["print angle_cos(nixon_v, rail_v)"], "collapsed": false, "prompt_number": 28, "metadata": {}, "outputs": [{"stream": "stdout", "text": ["0.847799936746\n"], "output_type": "stream"}]}], "metadata": {}}], "nbformat_minor": 0, "metadata": {"name": "", "signature": "sha256:f0dd2cf45e1be437a5fbf10b5d1cdc156db3cd5fa241cbe4b9b42892ed36bf94"}}