supervised mails.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import model_selection, preprocessing, linear_model, naive_bayes, metrics, svm\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer\n",
    "from sklearn import decomposition, ensemble\n",
    "\n",
    "import pandas, numpy, string\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# load the dataset\n",
    "data = open('E:\\irFinal\\supermails11.txt').read()\n",
    "dataz=data.split('\\t')\n",
    "labels, texts = [], []\n",
    "for line in enumerate(dataz):\n",
    "    linez=line[1].split('\\n')\n",
    "    #print(linez)\n",
    "    if linez[0]=='':\n",
    "        s=''\n",
    "        for i in range(1,len(linez)):\n",
    "            s+=linez[i]\n",
    "        if s!='':\n",
    "            texts.append(s)\n",
    "    else:\n",
    "        labels.append(linez[0])\n",
    "        s=''\n",
    "        for i in range(1,len(linez)):\n",
    "            s+=linez[i]\n",
    "        if s!='':\n",
    "            texts.append(s)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a dataframe using texts and lables\n",
    "trainDF = pandas.DataFrame()\n",
    "trainDF['text'] = texts\n",
    "trainDF['label'] = labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     \"TechGig Code Gladiators 2020 crosses 1 lakh r...\n",
       "1     \"Has being given a free Pro membership helped?...\n",
       "2     \"TechGig Code Gladiators 2020 crosses 1 lakh r...\n",
       "3     \"Are you a fresher? Here's why you should atte...\n",
       "4     \"The May Long Challenge is just around the cor...\n",
       "5     \"Live Problem DiscussionWe’ve been organizing ...\n",
       "6     Codecheffers are making the most of this time ...\n",
       "7     \"Note that Cook-Off is on a Monday this month ...\n",
       "8     \"You can book your spot here. To Register, jus...\n",
       "9     \"Hey Geek,   Are you ready to face the big sho...\n",
       "10    \"Women across different career fields are leav...\n",
       "11    \"You're receiving this monthly email because y...\n",
       "12    Your Google Account was just signed in to from...\n",
       "13    \"Adding Google Chrome, Google Chrome OS and Go...\n",
       "14    \"No changes to our Privacy Policy: We’re not m...\n",
       "15    Google can use this phone number to make sure ...\n",
       "16    \"No changes to our Privacy Policy: We’re not m...\n",
       "17    \"If you’re the guardian of a child under the a...\n",
       "18    \"You'll get Google sign-in prompts on all your...\n",
       "19    \"Now you can try it for yourself with free upg...\n",
       "20    \"Offer ends midnight GMT 30 June 2020. Don't m...\n",
       "21    \"If you have extra time or you'd like to take ...\n",
       "22    \"Plus, discover how to adapt your classes for ...\n",
       "23    \"This is it. In a matter of hours, the Unlimit...\n",
       "24    \"We hope that you're safe and well, and taking...\n",
       "25    \"What could an investment in a year of limitle...\n",
       "26    \"Alison is all set to launch a series of high-...\n",
       "27    \"Adobe Create    Learn new ways to create with...\n",
       "28    \"Think of the opportunities you could unlock w...\n",
       "29    \"We’ve already given Pro memberships to almost...\n",
       "30    \"A lot has happened on Facebook since you last...\n",
       "31    \"Instagram sends updates like this to help you...\n",
       "32    \" If you don't want to receive these emails fr...\n",
       "33    \"Let's get started with this new survey.Don't ...\n",
       "34    \"Ever wondered about who the worst ambassador ...\n",
       "35    \"He's a cover singing sensation. He's graced a...\n",
       "36    \"Gautam Jha invited you to like The Entertainm...\n",
       "37    \"She's a star in the making. Starring in Briti...\n",
       "38    \"Views flood his Instagram and YouTube covers....\n",
       "39    \"The Step-Up initiative is conceptualized with...\n",
       "40    \"Happy Thursday.This is a strange time. I'm tr...\n",
       "41    \"The architect Christopher Alexander on how to...\n",
       "42    \"Sharon Salzberg on faith and keeping an open ...\n",
       "43    \"1 QUESTION FOR YOUCan you sit still, do nothi...\n",
       "44    \"Happy Thursday! Allow me to spice up your inb...\n",
       "45    \"2 QUOTES FROM OTHERSI.The author and generall...\n",
       "46    \"Bonus quote:Bryson's description reminds me o...\n",
       "47    \"The poet and Pulitzer Prize winner Mary Olive...\n",
       "48    \"1 QUESTION FOR YOUWhat would 10-year-old me s...\n",
       "49    \"2 QUOTES FROM OTHERSI.E.O. Wilson, a sociobio...\n",
       "50    \"Another year complete. Here's a quick recap o...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trainDF.text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>\"TechGig Code Gladiators 2020 crosses 1 lakh r...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>\"Has being given a free Pro membership helped?...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"TechGig Code Gladiators 2020 crosses 1 lakh r...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>\"Are you a fresher? Here's why you should atte...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>\"The May Long Challenge is just around the cor...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>\"Live Problem DiscussionWe’ve been organizing ...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Codecheffers are making the most of this time ...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>\"Note that Cook-Off is on a Monday this month ...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>\"You can book your spot here. To Register, jus...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>\"Hey Geek,   Are you ready to face the big sho...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>\"Women across different career fields are leav...</td>\n",
       "      <td>competitions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>\"You're receiving this monthly email because y...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Your Google Account was just signed in to from...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>\"Adding Google Chrome, Google Chrome OS and Go...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>\"No changes to our Privacy Policy: We’re not m...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Google can use this phone number to make sure ...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>\"No changes to our Privacy Policy: We’re not m...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>\"If you’re the guardian of a child under the a...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>\"You'll get Google sign-in prompts on all your...</td>\n",
       "      <td>security</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>\"Now you can try it for yourself with free upg...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>\"Offer ends midnight GMT 30 June 2020. Don't m...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>\"If you have extra time or you'd like to take ...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>\"Plus, discover how to adapt your classes for ...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>\"This is it. In a matter of hours, the Unlimit...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>\"We hope that you're safe and well, and taking...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>\"What could an investment in a year of limitle...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>\"Alison is all set to launch a series of high-...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>\"Adobe Create    Learn new ways to create with...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>\"Think of the opportunities you could unlock w...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>\"We’ve already given Pro memberships to almost...</td>\n",
       "      <td>promotions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>\"A lot has happened on Facebook since you last...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>\"Instagram sends updates like this to help you...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>\" If you don't want to receive these emails fr...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>\"Let's get started with this new survey.Don't ...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>\"Ever wondered about who the worst ambassador ...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>\"He's a cover singing sensation. He's graced a...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>\"Gautam Jha invited you to like The Entertainm...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>\"She's a star in the making. Starring in Briti...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>\"Views flood his Instagram and YouTube covers....</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>\"The Step-Up initiative is conceptualized with...</td>\n",
       "      <td>social media</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>\"Happy Thursday.This is a strange time. I'm tr...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>\"The architect Christopher Alexander on how to...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>\"Sharon Salzberg on faith and keeping an open ...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>\"1 QUESTION FOR YOUCan you sit still, do nothi...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>\"Happy Thursday! Allow me to spice up your inb...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>\"2 QUOTES FROM OTHERSI.The author and generall...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>\"Bonus quote:Bryson's description reminds me o...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>\"The poet and Pulitzer Prize winner Mary Olive...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>\"1 QUESTION FOR YOUWhat would 10-year-old me s...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>\"2 QUOTES FROM OTHERSI.E.O. Wilson, a sociobio...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>\"Another year complete. Here's a quick recap o...</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 text         label\n",
       "0   \"TechGig Code Gladiators 2020 crosses 1 lakh r...  competitions\n",
       "1   \"Has being given a free Pro membership helped?...  competitions\n",
       "2   \"TechGig Code Gladiators 2020 crosses 1 lakh r...  competitions\n",
       "3   \"Are you a fresher? Here's why you should atte...  competitions\n",
       "4   \"The May Long Challenge is just around the cor...  competitions\n",
       "5   \"Live Problem DiscussionWe’ve been organizing ...  competitions\n",
       "6   Codecheffers are making the most of this time ...  competitions\n",
       "7   \"Note that Cook-Off is on a Monday this month ...  competitions\n",
       "8   \"You can book your spot here. To Register, jus...  competitions\n",
       "9   \"Hey Geek,   Are you ready to face the big sho...  competitions\n",
       "10  \"Women across different career fields are leav...  competitions\n",
       "11  \"You're receiving this monthly email because y...      security\n",
       "12  Your Google Account was just signed in to from...      security\n",
       "13  \"Adding Google Chrome, Google Chrome OS and Go...      security\n",
       "14  \"No changes to our Privacy Policy: We’re not m...      security\n",
       "15  Google can use this phone number to make sure ...      security\n",
       "16  \"No changes to our Privacy Policy: We’re not m...      security\n",
       "17  \"If you’re the guardian of a child under the a...      security\n",
       "18  \"You'll get Google sign-in prompts on all your...      security\n",
       "19  \"Now you can try it for yourself with free upg...    promotions\n",
       "20  \"Offer ends midnight GMT 30 June 2020. Don't m...    promotions\n",
       "21  \"If you have extra time or you'd like to take ...    promotions\n",
       "22  \"Plus, discover how to adapt your classes for ...    promotions\n",
       "23  \"This is it. In a matter of hours, the Unlimit...    promotions\n",
       "24  \"We hope that you're safe and well, and taking...    promotions\n",
       "25  \"What could an investment in a year of limitle...    promotions\n",
       "26  \"Alison is all set to launch a series of high-...    promotions\n",
       "27  \"Adobe Create    Learn new ways to create with...    promotions\n",
       "28  \"Think of the opportunities you could unlock w...    promotions\n",
       "29  \"We’ve already given Pro memberships to almost...    promotions\n",
       "30  \"A lot has happened on Facebook since you last...  social media\n",
       "31  \"Instagram sends updates like this to help you...  social media\n",
       "32  \" If you don't want to receive these emails fr...  social media\n",
       "33  \"Let's get started with this new survey.Don't ...  social media\n",
       "34  \"Ever wondered about who the worst ambassador ...  social media\n",
       "35  \"He's a cover singing sensation. He's graced a...  social media\n",
       "36  \"Gautam Jha invited you to like The Entertainm...  social media\n",
       "37  \"She's a star in the making. Starring in Briti...  social media\n",
       "38  \"Views flood his Instagram and YouTube covers....  social media\n",
       "39  \"The Step-Up initiative is conceptualized with...  social media\n",
       "40  \"Happy Thursday.This is a strange time. I'm tr...    motivation\n",
       "41  \"The architect Christopher Alexander on how to...    motivation\n",
       "42  \"Sharon Salzberg on faith and keeping an open ...    motivation\n",
       "43  \"1 QUESTION FOR YOUCan you sit still, do nothi...    motivation\n",
       "44  \"Happy Thursday! Allow me to spice up your inb...    motivation\n",
       "45  \"2 QUOTES FROM OTHERSI.The author and generall...    motivation\n",
       "46  \"Bonus quote:Bryson's description reminds me o...    motivation\n",
       "47  \"The poet and Pulitzer Prize winner Mary Olive...    motivation\n",
       "48  \"1 QUESTION FOR YOUWhat would 10-year-old me s...    motivation\n",
       "49  \"2 QUOTES FROM OTHERSI.E.O. Wilson, a sociobio...    motivation\n",
       "50  \"Another year complete. Here's a quick recap o...    motivation"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trainDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split the dataset into training and validation datasets \n",
    "train_x, valid_x, train_y, valid_y = model_selection.train_test_split(trainDF['text'], trainDF['label'])\n",
    "\n",
    "# label encode the target variable \n",
    "encoder = preprocessing.LabelEncoder()\n",
    "train_y = encoder.fit_transform(train_y)\n",
    "valid_y = encoder.fit_transform(valid_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "21    \"If you have extra time or you'd like to take ...\n",
       "27    \"Adobe Create    Learn new ways to create with...\n",
       "48    \"1 QUESTION FOR YOUWhat would 10-year-old me s...\n",
       "5     \"Live Problem DiscussionWe’ve been organizing ...\n",
       "50    \"Another year complete. Here's a quick recap o...\n",
       "43    \"1 QUESTION FOR YOUCan you sit still, do nothi...\n",
       "14    \"No changes to our Privacy Policy: We’re not m...\n",
       "28    \"Think of the opportunities you could unlock w...\n",
       "41    \"The architect Christopher Alexander on how to...\n",
       "10    \"Women across different career fields are leav...\n",
       "22    \"Plus, discover how to adapt your classes for ...\n",
       "1     \"Has being given a free Pro membership helped?...\n",
       "45    \"2 QUOTES FROM OTHERSI.The author and generall...\n",
       "13    \"Adding Google Chrome, Google Chrome OS and Go...\n",
       "44    \"Happy Thursday! Allow me to spice up your inb...\n",
       "4     \"The May Long Challenge is just around the cor...\n",
       "25    \"What could an investment in a year of limitle...\n",
       "18    \"You'll get Google sign-in prompts on all your...\n",
       "3     \"Are you a fresher? Here's why you should atte...\n",
       "17    \"If you’re the guardian of a child under the a...\n",
       "24    \"We hope that you're safe and well, and taking...\n",
       "31    \"Instagram sends updates like this to help you...\n",
       "34    \"Ever wondered about who the worst ambassador ...\n",
       "8     \"You can book your spot here. To Register, jus...\n",
       "15    Google can use this phone number to make sure ...\n",
       "6     Codecheffers are making the most of this time ...\n",
       "30    \"A lot has happened on Facebook since you last...\n",
       "47    \"The poet and Pulitzer Prize winner Mary Olive...\n",
       "29    \"We’ve already given Pro memberships to almost...\n",
       "40    \"Happy Thursday.This is a strange time. I'm tr...\n",
       "26    \"Alison is all set to launch a series of high-...\n",
       "9     \"Hey Geek,   Are you ready to face the big sho...\n",
       "11    \"You're receiving this monthly email because y...\n",
       "2     \"TechGig Code Gladiators 2020 crosses 1 lakh r...\n",
       "16    \"No changes to our Privacy Policy: We’re not m...\n",
       "49    \"2 QUOTES FROM OTHERSI.E.O. Wilson, a sociobio...\n",
       "23    \"This is it. In a matter of hours, the Unlimit...\n",
       "39    \"The Step-Up initiative is conceptualized with...\n",
       "Name: text, dtype: object"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# count vectors\n",
    "# create a count vectorizer object \n",
    "count_vect = CountVectorizer(analyzer='word', token_pattern=r'\\w{1,}')\n",
    "count_vect.fit(trainDF['text'])\n",
    "\n",
    "# transform the training and validation data using count vectorizer object\n",
    "xtrain_count =  count_vect.transform(train_x)\n",
    "xvalid_count =  count_vect.transform(valid_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# word level tf-idf\n",
    "tfidf_vect = TfidfVectorizer(analyzer='word', token_pattern=r'\\w{1,}', max_features=5000)\n",
    "tfidf_vect.fit(trainDF['text'])\n",
    "xtrain_tfidf =  tfidf_vect.transform(train_x)\n",
    "xvalid_tfidf =  tfidf_vect.transform(valid_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ngram level tf-idf \n",
    "tfidf_vect_ngram = TfidfVectorizer(analyzer='word', token_pattern=r'\\w{1,}', ngram_range=(2,3), max_features=5000)\n",
    "tfidf_vect_ngram.fit(trainDF['text'])\n",
    "xtrain_tfidf_ngram =  tfidf_vect_ngram.transform(train_x)\n",
    "xvalid_tfidf_ngram =  tfidf_vect_ngram.transform(valid_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_model(classifier, feature_vector_train, label, feature_vector_valid, is_neural_net=False):\n",
    "    # fit the training dataset on the classifier\n",
    "    classifier.fit(feature_vector_train, label)\n",
    "    \n",
    "    # predict the labels on validation dataset\n",
    "    predictions = classifier.predict(feature_vector_valid)\n",
    "    \n",
    "    if is_neural_net:\n",
    "        predictions = predictions.argmax(axis=-1)\n",
    "    \n",
    "    return metrics.accuracy_score(predictions, valid_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NB, Count Vectors:  0.46153846153846156\n"
     ]
    }
   ],
   "source": [
    "# Naive Bayes on Count Vectors\n",
    "nbcaccuracy = train_model(naive_bayes.MultinomialNB(), xtrain_count, train_y, xvalid_count)\n",
    "print(\"NB, Count Vectors: \", nbcaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NB, WordLevel TF-IDF:  0.46153846153846156\n"
     ]
    }
   ],
   "source": [
    "# Naive Bayes on Word Level TF IDF Vectors\n",
    "nbtaccuracy = train_model(naive_bayes.MultinomialNB(), xtrain_tfidf, train_y, xvalid_tfidf)\n",
    "print(\"NB, WordLevel TF-IDF: \", nbtaccuracy)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NB, N-Gram Vectors:  0.3076923076923077\n"
     ]
    }
   ],
   "source": [
    "nbnaccuracy = train_model(naive_bayes.MultinomialNB(), xtrain_tfidf_ngram, train_y, xvalid_tfidf_ngram)\n",
    "print(\"NB, N-Gram Vectors: \", nbnaccuracy)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM, TF-IDF:  0.3076923076923077\n"
     ]
    }
   ],
   "source": [
    "# SVM on Ngram Level TF IDF Vectors\n",
    "svmtaccuracy = train_model(svm.SVC(), xtrain_tfidf, train_y, xvalid_tfidf)\n",
    "print(\"SVM, TF-IDF: \", svmtaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM,Count Vectors:  0.38461538461538464\n"
     ]
    }
   ],
   "source": [
    "# SVM on Ngram Level TF IDF Vectors\n",
    "svmcaccuracy = train_model(svm.SVC(), xtrain_count, train_y, xvalid_count)\n",
    "print(\"SVM,Count Vectors: \", svmcaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM, N-Gram Vectors:  0.15384615384615385\n"
     ]
    }
   ],
   "source": [
    "# SVM on Ngram Level TF IDF Vectors\n",
    "svmnaccuracy = train_model(svm.SVC(), xtrain_tfidf_ngram, train_y, xvalid_tfidf_ngram)\n",
    "print(\"SVM, N-Gram Vectors: \", svmnaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Bagging Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\admin\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bagging Model, Count Vectors:  0.46153846153846156\n"
     ]
    }
   ],
   "source": [
    "# RF on Count Vectors\n",
    "rfcaccuracy = train_model(ensemble.RandomForestClassifier(), xtrain_count, train_y, xvalid_count)\n",
    "print(\"Bagging Model, Count Vectors: \", rfcaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bagging Model, WordLevel TF-IDF:  0.38461538461538464\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\admin\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# RF on Word Level TF IDF Vectors\n",
    "rftaccuracy = train_model(ensemble.RandomForestClassifier(), xtrain_tfidf, train_y, xvalid_tfidf)\n",
    "print(\"Bagging Model, WordLevel TF-IDF: \", rftaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bagging Model, N-Gram Vectors:  0.23076923076923078\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\admin\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# RF on Word Level TF IDF Vectors\n",
    "rfnaccuracy = train_model(ensemble.RandomForestClassifier(), xtrain_tfidf_ngram, train_y, xvalid_tfidf_ngram)\n",
    "print(\"Bagging Model, N-Gram Vectors: \", rfnaccuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.preprocessing import text, sequence\n",
    "from keras import layers, models, optimizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN, Count Vectors:  0.38461538461538464\n",
      "KNN, WordLevel TF-IDF:  0.5384615384615384\n",
      "KNN, N-Gram Vectors:  0.3076923076923077\n"
     ]
    }
   ],
   "source": [
    "#classification -knn - k_Nearest_Neighbours\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn=KNeighborsClassifier(n_neighbors=5)\n",
    "knn.fit(xtrain_count, train_y)\n",
    "knn.score(xvalid_count,valid_y)\n",
    "knnc=train_model(KNeighborsClassifier(n_neighbors=5),  xtrain_count, train_y, xvalid_count)\n",
    "knnt=train_model(KNeighborsClassifier(n_neighbors=5), xtrain_tfidf, train_y, xvalid_tfidf)\n",
    "knnn=train_model(KNeighborsClassifier(n_neighbors=5), xtrain_tfidf_ngram, train_y, xvalid_tfidf_ngram)\n",
    "print(\"KNN, Count Vectors: \", knnc)\n",
    "print(\"KNN, WordLevel TF-IDF: \", knnt)\n",
    "print(\"KNN, N-Gram Vectors: \", knnn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NB, Count Vectors:  0.46153846153846156\n",
      "NB, WordLevel TF-IDF:  0.46153846153846156\n",
      "NB, N-Gram Vectors:  0.3076923076923077\n",
      "SVM,Count Vectors:  0.38461538461538464\n",
      "SVM, TF-IDF:  0.3076923076923077\n",
      "SVM, N-Gram Vectors:  0.15384615384615385\n",
      "Bagging Model, Count Vectors:  0.46153846153846156\n",
      "Bagging Model, WordLevel TF-IDF:  0.38461538461538464\n",
      "Bagging Model, N-Gram Vectors:  0.23076923076923078\n",
      "KNN, Count Vectors:  0.38461538461538464\n",
      "KNN, WordLevel TF-IDF:  0.5384615384615384\n",
      "KNN, N-Gram Vectors:  0.3076923076923077\n"
     ]
    }
   ],
   "source": [
    "print(\"NB, Count Vectors: \", nbcaccuracy)\n",
    "print(\"NB, WordLevel TF-IDF: \", nbtaccuracy)\n",
    "print(\"NB, N-Gram Vectors: \", nbnaccuracy)\n",
    "print(\"SVM,Count Vectors: \", svmcaccuracy)\n",
    "print(\"SVM, TF-IDF: \", svmtaccuracy)\n",
    "print(\"SVM, N-Gram Vectors: \", svmnaccuracy)\n",
    "print(\"Bagging Model, Count Vectors: \", rfcaccuracy)\n",
    "print(\"Bagging Model, WordLevel TF-IDF: \", rftaccuracy)\n",
    "print(\"Bagging Model, N-Gram Vectors: \", rfnaccuracy)\n",
    "print(\"KNN, Count Vectors: \", knnc)\n",
    "print(\"KNN, WordLevel TF-IDF: \", knnt)\n",
    "print(\"KNN, N-Gram Vectors: \", knnn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Kmeans on our data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer, ENGLISH_STOP_WORDS\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.cluster import KMeans, MiniBatchKMeans\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.decomposition import TruncatedSVD \n",
    "from sklearn.preprocessing import normalize \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_raw_message(raw_message):\n",
    "    lines = raw_message.split('\\n')\n",
    "    email = {}\n",
    "    message = ''\n",
    "    keys_to_extract = ['from', 'to']\n",
    "    for line in lines:\n",
    "        if ':' not in line:\n",
    "            message += line.strip()\n",
    "            email['body'] = message\n",
    "        else:\n",
    "            pairs = line.split(':')\n",
    "            key = pairs[0].lower()\n",
    "            val = pairs[1].strip()\n",
    "            if key in keys_to_extract:\n",
    "                email[key] = val\n",
    "    return email\n",
    "\n",
    "def parse_into_emails(messages):\n",
    "    emails = [parse_raw_message(message) for message in messages]\n",
    "    return {\n",
    "        'body': map_to_list(emails, 'body'), \n",
    "        'to': map_to_list(emails, 'to'), \n",
    "        'from_': map_to_list(emails, 'from')\n",
    "    }\n",
    "\n",
    "def map_to_list(emails, key):\n",
    "    results = []\n",
    "    for email in emails:\n",
    "        if key not in email:\n",
    "            results.append('')\n",
    "        else:\n",
    "            results.append(email[key])\n",
    "    return results\n",
    "\n",
    "def top_tfidf_feats(row, features, top_n=20):\n",
    "    topn_ids = np.argsort(row)[::-1][:top_n]\n",
    "    top_feats = [(features[i], row[i]) for i in topn_ids]\n",
    "    df = pd.DataFrame(top_feats, columns=['features', 'score'])\n",
    "    return df\n",
    "\n",
    "def top_feats_in_doc(X, features, row_id, top_n=25):\n",
    "    row = np.squeeze(X[row_id].toarray())\n",
    "    return top_tfidf_feats(row, features, top_n)\n",
    "\n",
    "def top_mean_feats(X, features, grp_ids=None, min_tfidf=0.1, top_n=25):\n",
    "    if grp_ids:\n",
    "        D = X[grp_ids].toarray()\n",
    "    else:\n",
    "        D = X.toarray()\n",
    "\n",
    "    D[D < min_tfidf] = 0\n",
    "    tfidf_means = np.mean(D, axis=0)\n",
    "    return top_tfidf_feats(tfidf_means, features, top_n)\n",
    "\n",
    "def top_feats_per_cluster(X, y, features, min_tfidf=0.1, top_n=25):\n",
    "    dfs = []\n",
    "\n",
    "    labels = np.unique(y)\n",
    "    for label in labels:\n",
    "        ids = np.where(y==label) \n",
    "        feats_df = top_mean_feats(X, features, ids, min_tfidf=min_tfidf, top_n=top_n)\n",
    "        feats_df.label = label\n",
    "        dfs.append(feats_df)\n",
    "    return dfs\n",
    "\n",
    "def plot_tfidf_classfeats_h(dfs):\n",
    "    fig = plt.figure(figsize=(12, 9), facecolor=\"w\")\n",
    "    x = np.arange(len(dfs[0]))\n",
    "    for i, df in enumerate(dfs):\n",
    "        ax = fig.add_subplot(1, len(dfs), i+1)\n",
    "        ax.spines[\"top\"].set_visible(False)\n",
    "        ax.spines[\"right\"].set_visible(False)\n",
    "        ax.set_frame_on(False)\n",
    "        ax.get_xaxis().tick_bottom()\n",
    "        ax.get_yaxis().tick_left()\n",
    "        ax.set_xlabel(\"Tf-Idf Score\", labelpad=16, fontsize=14)\n",
    "        ax.set_title(\"cluster = \" + str(df.label), fontsize=16)\n",
    "        ax.ticklabel_format(axis='x', style='sci', scilimits=(-2,2))\n",
    "        ax.barh(x, df.score, align='center', color='#7530FF')\n",
    "        ax.set_yticks(x)\n",
    "        ax.set_ylim([-1, x[-1]+1])\n",
    "        yticks = ax.set_yticklabels(df.features)\n",
    "        plt.subplots_adjust(bottom=0.09, right=0.97, left=0.15, top=0.95, wspace=0.52)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     features     score\n",
      "0         pro  0.669148\n",
      "1    exciting  0.223049\n",
      "2        tips  0.223049\n",
      "3         did  0.223049\n",
      "4    starting  0.223049\n",
      "5       given  0.223049\n",
      "6         lot  0.223049\n",
      "7       going  0.223049\n",
      "8  codecademy  0.223049\n",
      "9        love  0.193475\n"
     ]
    }
   ],
   "source": [
    "stopwords = ENGLISH_STOP_WORDS.union(['ect', 'hou', 'com', 'recipient'])\n",
    "vect = TfidfVectorizer(analyzer='word', stop_words=stopwords, max_df=0.3, min_df=2)\n",
    "X = vect.fit_transform(trainDF.text)\n",
    "features = vect.get_feature_names()\n",
    "\n",
    "# Let's print the top 10 terms in document 1\n",
    "print(top_feats_in_doc(X, features, 1, 10))\n",
    "n_clusters = 5\n",
    "clf = KMeans(n_clusters=n_clusters, \n",
    "            max_iter=100, \n",
    "            init='k-means++', \n",
    "            n_init=1)\n",
    "labels = clf.fit_predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_tfidf_classfeats_h(top_feats_per_cluster(X, labels, features, 0.1, 25))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}