{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c6dd09fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dependencies and packages neededfrom sklearn import datasets\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sn; sn.set()\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.datasets import fetch_20newsgroups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "aebcecae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'alpha': 1.0, 'class_prior': None, 'fit_prior': True}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# this step is just to demonstrate the adjustable parameters within the algorithm and not a necessary step\n",
    "MultinomialNB().get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5f6a302f",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'fetch_20newsgroups' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Input \u001b[1;32mIn [1]\u001b[0m, in \u001b[0;36m<cell line: 2>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# import the dataset (stored in a variable)\u001b[39;00m\n\u001b[1;32m----> 2\u001b[0m data \u001b[38;5;241m=\u001b[39m \u001b[43mfetch_20newsgroups\u001b[49m()\n\u001b[0;32m      3\u001b[0m data\u001b[38;5;241m.\u001b[39mtarget_names\n",
      "\u001b[1;31mNameError\u001b[0m: name 'fetch_20newsgroups' is not defined"
     ]
    }
   ],
   "source": [
    "# import the dataset (stored in a variable)\n",
    "data = fetch_20newsgroups()\n",
    "data.target_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d61cee0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the categories\n",
    "categories = ['alt.atheism',\n",
    " 'comp.graphics',\n",
    " 'comp.os.ms-windows.misc',\n",
    " 'comp.sys.ibm.pc.hardware',\n",
    " 'comp.sys.mac.hardware',\n",
    " 'comp.windows.x',\n",
    " 'misc.forsale',\n",
    " 'rec.autos',\n",
    " 'rec.motorcycles',\n",
    " 'rec.sport.baseball',\n",
    " 'rec.sport.hockey',\n",
    " 'sci.crypt',\n",
    " 'sci.electronics',\n",
    " 'sci.med',\n",
    " 'sci.space',\n",
    " 'soc.religion.christian',\n",
    " 'talk.politics.guns',\n",
    " 'talk.politics.mideast',\n",
    " 'talk.politics.misc',\n",
    " 'talk.religion.misc']\n",
    "\n",
    "# Set training data to these categories\n",
    "train = fetch_20newsgroups(subset='train', categories = categories)\n",
    "# Set test data to these categories\n",
    "test = fetch_20newsgroups(subset='test', categories = categories)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "67f17ab2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "From: dfo@vttoulu.tko.vtt.fi (Foxvog Douglas)\n",
      "Subject: Re: Rewording the Second Amendment (ideas)\n",
      "Organization: VTT\n",
      "Lines: 58\n",
      "\n",
      "In article <1r1eu1$4t@transfer.stratus.com> cdt@sw.stratus.com (C. D. Tavares) writes:\n",
      ">In article <1993Apr20.083057.16899@ousrvr.oulu.fi>, dfo@vttoulu.tko.vtt.fi (Foxvog Douglas) writes:\n",
      ">> In article <1qv87v$4j3@transfer.stratus.com> cdt@sw.stratus.com (C. D. Tavares) writes:\n",
      ">> >In article <C5n3GI.F8F@ulowell.ulowell.edu>, jrutledg@cs.ulowell.edu (John Lawrence Rutledge) writes:\n",
      ">\n",
      ">> >> The massive destructive power of many modern weapons, makes the\n",
      ">> >> cost of an accidental or crimial usage of these weapons to great.\n",
      ">> >> The weapons of mass destruction need to be in the control of\n",
      ">> >> the government only.  Individual access would result in the\n",
      ">> >> needless deaths of millions.  This makes the right of the people\n",
      ">> >> to keep and bear many modern weapons non-existant.\n",
      "\n",
      ">> >Thanks for stating where you're coming from.  Needless to say, I\n",
      ">> >disagree on every count.\n",
      "\n",
      ">> You believe that individuals should have the right to own weapons of\n",
      ">> mass destruction?  I find it hard to believe that you would support a \n",
      ">> neighbor's right to keep nuclear weapons, biological weapons, and nerve\n",
      ">> gas on his/her property.  \n",
      "\n",
      ">> If we cannot even agree on keeping weapons of mass destruction out of\n",
      ">> the hands of individuals, can there be any hope for us?\n",
      "\n",
      ">I don't sign any blank checks.\n",
      "\n",
      "Of course.  The term must be rigidly defined in any bill.\n",
      "\n",
      ">When Doug Foxvog says \"weapons of mass destruction,\" he means CBW and\n",
      ">nukes.  When Sarah Brady says \"weapons of mass destruction\" she means\n",
      ">Street Sweeper shotguns and semi-automatic SKS rifles.  \n",
      "\n",
      "I doubt she uses this term for that.  You are using a quote allegedly\n",
      "from her, can you back it up?\n",
      "\n",
      ">When John\n",
      ">Lawrence Rutledge says \"weapons of mass destruction,\" and then immediately\n",
      ">follows it with:\n",
      "\n",
      ">>> The US has thousands of people killed each year by handguns,\n",
      ">>> this number can easily be reduced by putting reasonable restrictions\n",
      ">>> on them.\n",
      "\n",
      ">...what does Rutledge mean by the term?\n",
      "\n",
      "I read the article as presenting first an argument about weapons of mass\n",
      "destruction (as commonly understood) and then switching to other topics.\n",
      "The first point evidently was to show that not all weapons should be\n",
      "allowed, and then the later analysis was, given this understanding, to\n",
      "consider another class.\n",
      "\n",
      ">cdt@rocket.sw.stratus.com   --If you believe that I speak for my company,\n",
      ">OR cdt@vos.stratus.com        write today for my special Investors' Packet...\n",
      "\n",
      "\n",
      "\n",
      "-- \n",
      "doug foxvog\n",
      "douglas.foxvog@vtt.fi\n",
      "\n",
      "The amount of datapoints in the train data is:  11314\n"
     ]
    }
   ],
   "source": [
    "#Print the training data\n",
    "print(train.data[5])\n",
    "print('The amount of datapoints in the train data is: ',len(train.data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fcdfc062",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "From: banschbach@vms.ocom.okstate.edu\n",
      "Subject: Re: Candida(yeast) Bloom, Fact or Fiction\n",
      "Organization: OSU College of Osteopathic Medicine\n",
      "Lines: 91\n",
      "Nntp-Posting-Host: vms.ocom.okstate.edu\n",
      "\n",
      "In article <1rp8p1$2d3@usenet.INS.CWRU.Edu>, esd3@po.CWRU.Edu (Elisabeth S. Davidson) writes:\n",
      "> \n",
      "> In a previous article, banschbach@vms.ocom.okstate.edu () says:\n",
      ">>least a few \"enlightened\" physicians practicing in the U.S.  It's really \n",
      ">>too bad that most U.S. medical schools don't cover nutrition because if \n",
      ">>they did, candida would not be viewed as a non-disease by so many in the \n",
      ">>medical profession.\n",
      "> \n",
      "> Case Western Reserve Med School teaches nutrition in its own section as\n",
      "> well as covering it in other sections as they apply (i.e. B12\n",
      "> deficiency in neuro as a cause of neuropathy, B12 deficiency in\n",
      "> hematology as a cause of megaloblastic anemia), yet I sill\n",
      "> hold the viewpoint of mainstream medicine:  candida can cause\n",
      "> mucocutaneous candidiasis, and, in already very sick patients\n",
      "> with damaged immune systems like AIDS and cancer patients,\n",
      "> systemic candida infection.  I think \"The Yeast Connection\" is\n",
      "> a bunch of hooey.  What does this have to do with how well\n",
      "> nutrition is taught, anyway?\n",
      "\n",
      "Elisabeth, let's set the record straight for the nth time, I have not read \n",
      "\"The Yeast Connection\".  So anything that I say is not due to brainwashing \n",
      "by this \"hated\" book.  It's okay I guess to hate the book, by why hate me?\n",
      "Elisabeth, I'm going to quote from Zinsser's Microbiology, 20th Edition.\n",
      "A book that you should be familiar with and not \"hate\". \"Candida species \n",
      "colonize the mucosal surfaces of all humans during birth or shortly \n",
      "thereafter.  The risk of endogenous infection is clearly ever present.  \n",
      "Indeed, candidiasis occurs worldwide and is the most common systemic \n",
      "mycosis.\"  Neutrophils play the main role in preventing a systemic \n",
      "infection(candidiasis) so you would have to have a low neutrophil count or \n",
      "\"sick\" neutrophils to see a systemic infection.  Poor diet and persistent \n",
      "parasitic infestation set many third world residents up for candidiasis.\n",
      "Your assessment of candidiasis in the U.S. is correct and I do not dispute \n",
      "it.\n",
      "\n",
      "What I posted was a discussion of candida blooms, without systemic \n",
      "infection.  These blooms would be responsible for local sites of irritation\n",
      "(GI tract, mouth, vagina and sinus cavity).  Knocking down the bacterial \n",
      "competition for candida was proposed as a possible trigger for candida \n",
      "blooms.  Let me quote from Zinsser's again: \"However, some factors, such as \n",
      "the use of a broad-spectrum antibacterial antibiotic, may predispose to \n",
      "both mucosal and systemic infections\".  I was addressing mucosal infections\n",
      "(I like the term blooms better).  The nutrition course that I teach covers \n",
      "this effect of antibiotic treatment as well as the \"cure\".  I guess that \n",
      "your nutrition course does not, too bad.  \n",
      "\n",
      "\n",
      ">>Here is a brief primer on yeast.  Yeast infections, as they are commonly \n",
      ">>called, are not truely caused by yeasts.  The most common organism responsible\n",
      ">>for this type of infection is Candida albicans or Monilia which is actually a \n",
      ">>yeast-like fungus.  \n",
      "> \n",
      "> Well, maybe I'm getting picky, but I always thought that a yeast\n",
      "> was one form that a fungus could exist in, the other being the\n",
      "> mold form.  Many fungi can occur as either yeasts or molds, \n",
      "> depending on environment.  Candida exibits what is known as\n",
      "> reverse dimorphism - it exists as a mold in the tissues\n",
      "> but exists as a yeast in the environment.  Should we maybe\n",
      "> call it a mold infection?  a fungus infection?  Maybe we\n",
      "> should say it is caused by a mold-like fungus.\n",
      ">  \n",
      ">> \n",
      ">>Martin Banschbach, Ph.D.\n",
      ">>Professor of Biochemistry and Chairman\n",
      ">>Department of Biochemistry and Microbiology\n",
      ">>OSU College of Osteopathic Medicine\n",
      ">>1111 West 17th St.\n",
      ">>Tulsa, Ok. 74107\n",
      ">>\n",
      "> \n",
      "> You're the chairman of Biochem and Micro and you didn't know \n",
      "> that a yeast is a form of a fungus?  (shudder)\n",
      "> Or maybe you did know, and were oversimplifying?\n",
      "\n",
      "My, my Elisabeth, do I detect a little of Steve Dyer in you?  If you \n",
      "noticed my faculty rank, I'm a biochemist, not a microbiologist.\n",
      "Candida is classifed as a fungus(according to Zinsser's).  But, as you point \n",
      "out, it displays dimorphism.  It is capable of producing yeast cells, \n",
      "pseudohyphae and true hyphae.  Elisabeth, you are probably a microbiologist \n",
      "and that makes a lot of sense to you.  To a biochemist, it's a lot of \n",
      "Greek.  So I called it a yeast-like fungus, go ahead and crucify me.\n",
      "\n",
      "You know Elisabeth, I still haven't been able to figure out why such a small \n",
      "little organism like Candida can bring out so much hostility in people in \n",
      "Sci. Med.  And I must admitt that I got sucked into the mud slinging too.\n",
      "I keep hoping that if people will just take the time to think about what \n",
      "I've said, that it will make sense.  I'm not asking anyone here to buy into \n",
      "\"The Yeast Connection\" book because I don't know what's in that book, plain \n",
      "and simple. And to be honest with you, I'm beginning to wish that it was never \n",
      "written.\n",
      "\n",
      "Marty B.\n",
      "\n",
      "The amount of datapoints in the test data is:  7532\n"
     ]
    }
   ],
   "source": [
    "#Print the test data\n",
    "print(test.data[5])\n",
    "print('The amount of datapoints in the test data is: ',len(test.data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "80acdb7c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 7, 11,  0, ...,  9,  3, 15], dtype=int64)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a model based on Multinomial Naive Bayes\n",
    "model = make_pipeline(TfidfVectorizer(), MultinomialNB())\n",
    "# Fit the data to the model to train it\n",
    "model.fit(train.data, train.target)\n",
    "# Create the labels by classifying the test data\n",
    "labels = model.predict(test.data)\n",
    "labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "79d601bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 7,  5,  0, ...,  9,  6, 15], dtype=int64)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7e1c3c00",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[166   0   0   1   0   1   0   0   1   1   1   3   0   6   3 123   4   8\n",
      "    0   1]\n",
      " [  1 252  15  12   9  18   1   2   1   5   2  41   4   0   6  15   4   1\n",
      "    0   0]\n",
      " [  0  14 258  45   3   9   0   2   1   3   2  25   1   0   6  23   2   0\n",
      "    0   0]\n",
      " [  0   5  11 305  17   1   3   6   1   0   2  19  13   0   5   3   1   0\n",
      "    0   0]\n",
      " [  0   3   8  23 298   0   3   8   1   3   1  16   8   0   2   8   3   0\n",
      "    0   0]\n",
      " [  1  21  17  13   2 298   1   0   1   1   0  23   0   1   4  10   2   0\n",
      "    0   0]\n",
      " [  0   1   3  31  12   1 271  19   4   4   6   5  12   6   3   9   3   0\n",
      "    0   0]\n",
      " [  0   1   0   3   0   0   4 364   3   2   2   4   1   1   3   3   4   0\n",
      "    1   0]\n",
      " [  0   0   0   1   0   0   2  10 371   0   0   4   0   0   0   8   2   0\n",
      "    0   0]\n",
      " [  0   0   0   0   1   0   0   4   0 357  22   0   0   0   2   9   1   1\n",
      "    0   0]\n",
      " [  0   0   0   0   0   0   0   1   0   4 387   1   0   0   1   5   0   0\n",
      "    0   0]\n",
      " [  0   2   1   0   0   1   1   3   0   0   0 383   1   0   0   3   1   0\n",
      "    0   0]\n",
      " [  0   4   2  17   5   0   2   8   7   1   2  78 235   3  11  15   2   1\n",
      "    0   0]\n",
      " [  2   3   0   1   1   3   1   0   2   3   4  11   5 292   6  52   6   4\n",
      "    0   0]\n",
      " [  0   2   0   1   0   3   0   2   1   0   1   6   1   2 351  19   4   0\n",
      "    1   0]\n",
      " [  2   0   0   0   0   0   0   0   1   0   0   0   0   1   2 392   0   0\n",
      "    0   0]\n",
      " [  0   0   0   1   0   0   2   0   1   1   0  10   0   0   1   6 341   1\n",
      "    0   0]\n",
      " [  0   1   0   0   0   0   0   0   0   1   0   2   0   0   0  24   3 344\n",
      "    1   0]\n",
      " [  2   0   0   0   0   0   0   1   0   0   1  11   0   1   7  35 118   5\n",
      "  129   0]\n",
      " [ 33   2   0   0   0   0   0   0   0   1   1   3   0   4   4 131  29   5\n",
      "    3  35]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.802     0.520     0.631       319\n",
      "           1      0.810     0.648     0.720       389\n",
      "           2      0.819     0.655     0.728       394\n",
      "           3      0.672     0.778     0.721       392\n",
      "           4      0.856     0.774     0.813       385\n",
      "           5      0.890     0.754     0.816       395\n",
      "           6      0.931     0.695     0.796       390\n",
      "           7      0.847     0.919     0.881       396\n",
      "           8      0.937     0.932     0.935       398\n",
      "           9      0.922     0.899     0.911       397\n",
      "          10      0.892     0.970     0.929       399\n",
      "          11      0.594     0.967     0.736       396\n",
      "          12      0.836     0.598     0.697       393\n",
      "          13      0.921     0.737     0.819       396\n",
      "          14      0.842     0.891     0.866       394\n",
      "          15      0.439     0.985     0.607       398\n",
      "          16      0.643     0.937     0.763       364\n",
      "          17      0.930     0.915     0.922       376\n",
      "          18      0.956     0.416     0.580       310\n",
      "          19      0.972     0.139     0.244       251\n",
      "\n",
      "    accuracy                          0.774      7532\n",
      "   macro avg      0.826     0.757     0.756      7532\n",
      "weighted avg      0.822     0.774     0.768      7532\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# display the metrics of the model's classification through confusion matrix and classification report\n",
    "# (accuracy / precision / recall / f1)\n",
    "from sklearn import metrics\n",
    "print(metrics.confusion_matrix(test.target, labels))\n",
    "\n",
    "\n",
    "print(metrics.classification_report(test.target, labels, digits=3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "fde1a47f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      Actual  Predicted\n",
      "0          7          7\n",
      "1          5         11\n",
      "2          0          0\n",
      "3         17         17\n",
      "4         19          0\n",
      "...      ...        ...\n",
      "7527      14         14\n",
      "7528       4          9\n",
      "7529       9          9\n",
      "7530       6          3\n",
      "7531      15         15\n",
      "\n",
      "[7532 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "# use Pandas to create a dataframe to view the ground truth and classifications side by side\n",
    "import pandas as pd\n",
    "data = {'Actual': test.target,\n",
    "        'Predicted': labels}\n",
    "df = pd.DataFrame(data)\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0bdd2d22",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(89.133125, 0.5, 'Predicted')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use Seaborn to create a confusion matrix heatmap and MatPlotLib to display\n",
    "confusion_matrix = metrics.confusion_matrix(test.target, labels)\n",
    "sn.heatmap(confusion_matrix.T, square=True, annot=True, fmt='d', cbar=False, xticklabels=train.target_names, yticklabels=train.target_names)\n",
    "\n",
    "plt.xlabel('Ground Truth')\n",
    "plt.ylabel('Predicted')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "af7ec109",
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a function to use the model\n",
    "def predict_text_category(s, train=train, model=model):\n",
    "    predict = model.predict([s])\n",
    "    return categories[predict[0]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "77fed1b0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'sci.space'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_text_category('Sending a shuttle to the moon')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ace2520",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}