4_naive_bayes_in_textblob.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Text Classification with TextBlob\n",
    "--------\n",
    "\n",
    "Let's experiment with an existing library for classification: [TextBlob](https://textblob.readthedocs.org/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training\n",
    "-----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "reset -fs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "from textblob.classifiers import NaiveBayesClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = [(\"🐈 🐯 🐱 🐩 🐱\",                   'cat') ,\n",
    "         (\"🐶 🐶 🐈 🐶 🐩 🐈 🐶 🐶\",         'dog'),\n",
    "         (\"🐈 🐈 🐯 🐶 🐈\",                   'cat'),\n",
    "         (\"🐈 🐈 🐈\",                         'cat'),\n",
    "         (\"🐶 🐶 🐯 🐈 🐩 🐱 🐩 🐶 🐩 🐶 \", 'dog'),\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "cl = NaiveBayesClassifier(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'cat'"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cl.classify(\"🐱\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'cat'"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cl.classify(\"🐶 🐶\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'dog'"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cl.classify(\"🐶 🐱\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'dog'"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cl.classify(\"🐈 🐈 🐶 🐶 🐩 🐯 🐯\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'cat'"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cl.classify(\"🐬\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Most Informative Features\n",
      "             contains(🐶) = True              dog : cat    =      2.2 : 1.0\n",
      "             contains(🐩) = True              dog : cat    =      2.2 : 1.0\n",
      "             contains(🐯) = False             dog : cat    =      1.3 : 1.0\n",
      "             contains(🐱) = True              dog : cat    =      1.3 : 1.0\n",
      "             contains(🐯) = True              cat : dog    =      1.2 : 1.0\n",
      "             contains(🐱) = False             cat : dog    =      1.2 : 1.0\n",
      "             contains(🐈) = True              dog : cat    =      1.0 : 1.0\n"
     ]
    }
   ],
   "source": [
    "cl.show_informative_features()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br> \n",
    "<br>\n",
    "\n",
    "----"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Textblob with words\n",
    "-----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = [\n",
    "    ('I love this sandwich.', 'pos'),\n",
    "    ('This is an amazing place!', 'pos'),\n",
    "    ('I feel very good about these beers.', 'pos'),\n",
    "    ('This is my best work.', 'pos'),\n",
    "    (\"What an awesome view\", 'pos'),\n",
    "    ('I do not like this restaurant', 'neg'),\n",
    "    ('I am tired of this stuff.', 'neg'),\n",
    "    (\"I can't deal with this\", 'neg'),\n",
    "    ('He is my sworn enemy!', 'neg'),\n",
    "    ('My boss is horrible.', 'neg')\n",
    "]\n",
    "test = [\n",
    "    ('The beer was good.', 'pos'),\n",
    "    ('I do not enjoy my job', 'neg'),\n",
    "    (\"I ain't feeling dandy today.\", 'neg'),\n",
    "    (\"I feel amazing!\", 'pos'),\n",
    "    ('Gary is a friend of mine.', 'pos'),\n",
    "    (\"I can't believe I'm doing this.\", 'neg')\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "cl = NaiveBayesClassifier(train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How do we measure success for Text Classification?\n",
    "--------\n",
    " \n",
    "$$Accuracy = \\frac{Correct}{Total}$$\n",
    "\n",
    "Fraction of docs classified correctly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.83\n"
     ]
    }
   ],
   "source": [
    "print(f\"{cl.accuracy(test):.2}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The beer was good.\n",
      "  observed:   pos\n",
      "  predicted:  pos\n",
      "\n",
      "I do not enjoy my job\n",
      "  observed:   neg\n",
      "  predicted:  neg\n",
      "\n",
      "I ain't feeling dandy today.\n",
      "  observed:   neg\n",
      "  predicted:  neg\n",
      "\n",
      "I feel amazing!\n",
      "  observed:   pos\n",
      "  predicted:  pos\n",
      "\n",
      "Gary is a friend of mine.\n",
      "  observed:   pos\n",
      "  predicted:  neg\n",
      "\n",
      "I can't believe I'm doing this.\n",
      "  observed:   neg\n",
      "  predicted:  neg\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for review, label in test:\n",
    "    print(review)\n",
    "    print(\"  observed:  \", label)\n",
    "    print(\"  predicted: \", cl.classify(review), end=\"\\n\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Confusion Matrix\n",
    "------\n",
    "![](images/confusion_matrix.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'pos'"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's see Naive Bayes ability to generalize\n",
    "cl.classify(\"Their burgers are amazing\")  #=> \"pos\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'neg'"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cl.classify(\"I don't like their pizza.\") #=> neg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Most Informative Features\n",
      "          contains(this) = True              neg : pos    =      2.3 : 1.0\n",
      "          contains(this) = False             pos : neg    =      1.8 : 1.0\n",
      "            contains(an) = False             neg : pos    =      1.6 : 1.0\n",
      "          contains(This) = False             neg : pos    =      1.6 : 1.0\n",
      "             contains(I) = False             pos : neg    =      1.4 : 1.0\n",
      "             contains(I) = True              neg : pos    =      1.4 : 1.0\n",
      "         contains(beers) = False             neg : pos    =      1.2 : 1.0\n",
      "           contains(n't) = False             pos : neg    =      1.2 : 1.0\n",
      "          contains(love) = False             neg : pos    =      1.2 : 1.0\n",
      "       contains(amazing) = False             neg : pos    =      1.2 : 1.0\n"
     ]
    }
   ],
   "source": [
    "cl.show_informative_features()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br> \n",
    "<br>\n",
    "\n",
    "----"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}