Spam_Url_Analysis_With_ML.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "1236b337",
   "metadata": {},
   "source": [
    "# Spam Url Analysis With Machine Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e6442584",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn import tree\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2da2f26a",
   "metadata": {},
   "outputs": [],
   "source": [
    "url = pd.read_csv('../Spam-Email-and-Url-Detection-Using-Machine-Learning/data-set/url_spam_classification.csv') # data set read"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "474af51e",
   "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>url</th>\n",
       "      <th>is_spam</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>https://briefingday.us8.list-manage.com/unsubs...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>https://www.hvper.com/</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>https://briefingday.com/m/v4n3i4f3</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>https://briefingday.com/n/20200618/m#commentform</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>https://briefingday.com/fan</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148298</th>\n",
       "      <td>https://cdn.substack.com/image/fetch/f_auto,q_...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148299</th>\n",
       "      <td>https://numlock.substack.com/subscribe</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148300</th>\n",
       "      <td>https://docs.google.com/forms/d/e/1FAIpQLSfDLO...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148301</th>\n",
       "      <td>https://bookclub.substack.com/</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148302</th>\n",
       "      <td>https://awards.substack.com/</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>148303 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      url  is_spam\n",
       "0       https://briefingday.us8.list-manage.com/unsubs...     True\n",
       "1                                  https://www.hvper.com/     True\n",
       "2                      https://briefingday.com/m/v4n3i4f3     True\n",
       "3        https://briefingday.com/n/20200618/m#commentform    False\n",
       "4                             https://briefingday.com/fan     True\n",
       "...                                                   ...      ...\n",
       "148298  https://cdn.substack.com/image/fetch/f_auto,q_...     True\n",
       "148299             https://numlock.substack.com/subscribe     True\n",
       "148300  https://docs.google.com/forms/d/e/1FAIpQLSfDLO...     True\n",
       "148301                     https://bookclub.substack.com/     True\n",
       "148302                       https://awards.substack.com/     True\n",
       "\n",
       "[148303 rows x 2 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "36e8a5e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(148303, 2)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "af37a83b",
   "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>url</th>\n",
       "      <th>is_spam</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>148303</td>\n",
       "      <td>148303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>87581</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>https://www.bloomberg.com/tosv2.html</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1086</td>\n",
       "      <td>101021</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         url is_spam\n",
       "count                                 148303  148303\n",
       "unique                                 87581       2\n",
       "top     https://www.bloomberg.com/tosv2.html   False\n",
       "freq                                    1086  101021"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "05e2dde2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_spam\n",
       "False    101021\n",
       "True      47282\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.groupby(url['is_spam']).size()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86b9bdbe",
   "metadata": {},
   "source": [
    "We have a total of 148.303 data. There are 101.021 safe and 47.282 spam."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "323a9d81",
   "metadata": {},
   "outputs": [],
   "source": [
    "url['is_spam'] = url.is_spam.apply(str) # Convert boolean column to string column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c39ff65b",
   "metadata": {},
   "outputs": [],
   "source": [
    "url['is_spam'] = url['is_spam'].apply(lambda x : 1 if x == \"True\" in x else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "12a4320f",
   "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>url</th>\n",
       "      <th>is_spam</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>https://briefingday.us8.list-manage.com/unsubs...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>https://www.hvper.com/</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>https://briefingday.com/m/v4n3i4f3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>https://briefingday.com/n/20200618/m#commentform</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>https://briefingday.com/fan</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148298</th>\n",
       "      <td>https://cdn.substack.com/image/fetch/f_auto,q_...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148299</th>\n",
       "      <td>https://numlock.substack.com/subscribe</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148300</th>\n",
       "      <td>https://docs.google.com/forms/d/e/1FAIpQLSfDLO...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148301</th>\n",
       "      <td>https://bookclub.substack.com/</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148302</th>\n",
       "      <td>https://awards.substack.com/</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>148303 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      url  is_spam\n",
       "0       https://briefingday.us8.list-manage.com/unsubs...        1\n",
       "1                                  https://www.hvper.com/        1\n",
       "2                      https://briefingday.com/m/v4n3i4f3        1\n",
       "3        https://briefingday.com/n/20200618/m#commentform        0\n",
       "4                             https://briefingday.com/fan        1\n",
       "...                                                   ...      ...\n",
       "148298  https://cdn.substack.com/image/fetch/f_auto,q_...        1\n",
       "148299             https://numlock.substack.com/subscribe        1\n",
       "148300  https://docs.google.com/forms/d/e/1FAIpQLSfDLO...        1\n",
       "148301                     https://bookclub.substack.com/        1\n",
       "148302                       https://awards.substack.com/        1\n",
       "\n",
       "[148303 rows x 2 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fce25255",
   "metadata": {},
   "outputs": [],
   "source": [
    "urls = url.iloc[:,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b7b798cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    https://briefingday.us8.list-manage.com/unsubs...\n",
       "1                               https://www.hvper.com/\n",
       "2                   https://briefingday.com/m/v4n3i4f3\n",
       "3     https://briefingday.com/n/20200618/m#commentform\n",
       "4                          https://briefingday.com/fan\n",
       "Name: url, dtype: object"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urls.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2e18281c",
   "metadata": {},
   "outputs": [],
   "source": [
    "ifSpam = url.iloc[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "45f7f040",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    1\n",
       "2    1\n",
       "3    0\n",
       "4    1\n",
       "Name: is_spam, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ifSpam.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0aa0e5b3",
   "metadata": {},
   "source": [
    "With CountVectorizer, text is analyzed and word counts are made and these are converted into vectors."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1f803bb",
   "metadata": {},
   "source": [
    "In this analysis, we will use CountVectorizer, but first, since the url structure is not like normal text, we will convert the urls into a format that CountVectorizer can process with a method called tokenizer."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f088989a",
   "metadata": {},
   "source": [
    "### Tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "2fa09e39",
   "metadata": {},
   "outputs": [],
   "source": [
    "def extractUrl(data):\n",
    "    url = str(data)\n",
    "    extractSlash = url.split('/')\n",
    "    result = []\n",
    "    \n",
    "    for i in extractSlash:\n",
    "        extractDash = str(i).split('-')\n",
    "        dotExtract = []\n",
    "        \n",
    "        for j in range(0,len(extractDash)):\n",
    "            extractDot = str(extractDash[j]).split('.')\n",
    "            dotExtract += extractDot\n",
    "            \n",
    "        result += extractDash + dotExtract\n",
    "    result = list(set(result))\n",
    "\n",
    "    return result\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe12326c",
   "metadata": {},
   "source": [
    "We have created the function that will perform the Tokenizer, the work of this function is as in the example below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "76099c05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['', 'net', 'emrahyldrm', 'http:', 'www.emrahyldrm', 'www', '1234.net', '1234']\n"
     ]
    }
   ],
   "source": [
    "example = \"http://www.emrahyldrm-1234.net\"\n",
    "a = extractUrl(example)\n",
    "print(a)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "084b5f84",
   "metadata": {},
   "source": [
    "We will use 75% of our dataset for training and 25% for testing : "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "95db0aa8",
   "metadata": {},
   "outputs": [],
   "source": [
    "urls_train, urls_test, ifSpam_train, ifSpam_test = train_test_split(urls, ifSpam, test_size=0.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5d1ed4ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "cv = CountVectorizer(tokenizer=extractUrl) # We defined the tokenizer in CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "454d0766",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = cv.fit_transform(urls_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8da2831a",
   "metadata": {},
   "outputs": [],
   "source": [
    "features_test = cv.transform(urls_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "affe795d",
   "metadata": {},
   "source": [
    "# Learning and Predicts "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4709dd1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "dtModel = tree.DecisionTreeClassifier() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "86452ad7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier()"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dtModel.fit(features, ifSpam_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "cfc3049e",
   "metadata": {},
   "outputs": [],
   "source": [
    "dtPredict = dtModel.predict(features_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "cb86727d",
   "metadata": {},
   "outputs": [],
   "source": [
    "lsvcModel = LinearSVC()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f8324ab4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearSVC()"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lsvcModel.fit(features, ifSpam_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9632b34e",
   "metadata": {},
   "outputs": [],
   "source": [
    "lsvcPredict = lsvcModel.predict(features_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "54f8660c",
   "metadata": {},
   "outputs": [],
   "source": [
    "sgdcModel = SGDClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "48933159",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SGDClassifier()"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgdcModel.fit(features, ifSpam_train) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2b3d14e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "sgdcPredict = sgdcModel.predict(features_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "47e539bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "nbModel = MultinomialNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "41185b7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MultinomialNB()"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbModel.fit(features, ifSpam_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "0328b7cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "nbPredict = nbModel.predict(features_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6fb5a3e",
   "metadata": {},
   "source": [
    "# Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "749f8b92",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import plot_confusion_matrix,plot_precision_recall_curve,plot_roc_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "4109a447",
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualization(model):\n",
    "    predict = model.predict(features_test)\n",
    "    plot_confusion_matrix(model,features_test,ifSpam_test)\n",
    "    plot_precision_recall_curve(model,features_test,ifSpam_test)\n",
    "    plot_roc_curve(model,features_test,ifSpam_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0382e003",
   "metadata": {},
   "source": [
    "# Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17ae4a73",
   "metadata": {},
   "source": [
    "### Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "849936d1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mislabeled out of a total of 37076 test entries: 444\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of mislabeled out of a total of %d test entries: %d\" % (urls_test.shape[0], \n",
    "                                                                      (ifSpam_test != dtPredict).sum()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "4a846b77",
   "metadata": {},
   "outputs": [],
   "source": [
    "successRate = 100.0 * f1_score(ifSpam_test, dtPredict, average='micro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "08703b06",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Success Rate was calculated as % : 98.80245981227749 with Decision Tree\n"
     ]
    }
   ],
   "source": [
    "print(\"The Success Rate was calculated as % : \" + str(successRate) + \" with Decision Tree\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "7b94a8da",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEGCAYAAADyuIefAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAdUklEQVR4nO3deZwdVZ338c+3O52FkH0zZCEBQiAgm5EEGHhYFBLEYZFRFiEPiwEFRWUUdMQoDCM+KIxRBEEzgCKIIg8BAyFGkEWWBGRJgjEhCWQlezp70t2/+eNWdyqhl3uTvn27b3/fr1e9uu65p6p+1YFfn1On6pQiAjMzyygpdABmZs2Jk6KZWYqToplZipOimVmKk6KZWUqbQgeQ1rN7aQwaUFboMCwH/3x7r0KHYDnYEhvZFlu1J/s47aSOsWp1ZVZ1X3tr6+SIGLUnx2tqzSopDhpQxquTBxQ6DMvBqIHDCx2C5eDlisl7vI+Vqyt5ZXL/rOqW9X235x4fsIk1q6RoZi1BUBlVhQ4ib5wUzSwnAVRRvA99OCmaWc6qcEvRzAyAINju7rOZWUYAle4+m5nt4GuKZmaJACqLeHYtJ0Uzy1nxXlF0UjSzHAXha4pmZtUiYHvx5kQnRTPLlahkjx6fbtacFM0sJwFUuaVoZraDW4pmZonMzdtOimZmQCYpbo/inZ/aSdHMchKIyiKetN9J0cxyVhXuPpuZAb6maGa2C1Hpa4pmZhmZmbedFM3MAIgQ26K00GHkjZOimeWsytcUzcwyMgMt7j6bmSU80GJmVsMDLWZmu6j0zdtmZhmB2B7FmzqK98zMLC880GJmlhLI3WczszQPtJiZJSLwLTlmZtUyAy1+zM/MrIYHWszMEoE8yayZWZpbimZmicx7n50UzcwS8usIzMyqZV5xWryjz8XbBjazvIgQVVGS1VIfSQMkPSNplqSZkq5JyrtLmiJpTvKzW1IuSeMlzZX0lqSjUvsak9SfI2lMqvxjkt5OthkvqcEmrpOimeWsMkqyWhpQAVwbEcOAkcBVkoYB1wNTI2IIMDX5DDAaGJIsY4E7IZNEgXHACOBoYFx1Ik3qfCG13aiGgnJSNLOcZOZTVFZLvfuJWBoRryfr64F3gH7AmcB9SbX7gLOS9TOB+yPjZaCrpL7AacCUiFgdEWuAKcCo5LvOEfFyRARwf2pfdfI1RTPLUU4zb/eUND31+e6IuPtDe5QGAUcCrwB9ImJp8tUyoE+y3g9YmNpsUVJWX/miWsrr5aRoZjnJ3JKT9ejzyogYXl8FSXsDjwBfjYjy9GW/iAhJsbux7g53n80sJ9XPPmezNERSGZmE+EBE/DEp/iDp+pL8XJ6ULwYGpDbvn5TVV96/lvJ6OSmaWc6qKMlqqU8yEvwr4J2IuC311USgegR5DPBYqvziZBR6JLAu6WZPBk6V1C0ZYDkVmJx8Vy5pZHKsi1P7qpO7z2aWk8zUYY1y8/ZxwEXA25LeSMq+DdwCPCzpMuA94LPJd5OA04G5wCbgkkw8sVrSTcC0pN6NEbE6Wf8ScC/QAXgyWerlpGhmOWuMCSEi4gWoc4j6lFrqB3BVHfuaAEyopXw6cGgucTkpmllOMrPkFO+VNydFM8tJ5jE/J8VWb/niMm69ZiBrV5SBgtM/v4qzL19Z8/0f7urFPTf24+G336ZLj0o2lpfww6v3ZfmStlRWwLlXruC081bzwaIybrx0MFVVoqICzrx0JWdcvAqAZx/rykPj+1BZCSM+Uc7l31laVzi2B8raVfGj38+mrG1Q2iZ4flI3fnPbPhxxXDmXf3sRKoEtm0r40dcHsfS99vTut5Wv/eg9unavYP3aUv7fNYNZuaxtoU+jgNxS3G2SRgE/AUqBX0bELfk8Xj6VtgnGfncJQw7bzKYNJVw96kCOOmE9+x64leWLy3j9r53o3W9bTf2J9/Zk4IFbuPH++axdVcplxx/MyeesoXvvCm5/fA5t2wWbN5ZwxUkHccyp6yhrG/zypn342eTZdO1Rya3XDOTvz+/NkcdvKOBZF6ftW8V15x3Ilk2llLYJfvzIP5j+TGeuvvl9vn/5/iyc24EzLlrOBV9Zxo+vHcQXvrOIqY/04M9/6MHhx5ZzyfWLufWrgwt9GgXV0NMqLVne0r2kUuAOMs8rDgPOT55rbJF69KlgyGGbAdhr7yoGHLCVlUvLAPjF9/px2XeWkH7UXILNG0uJgC0bS+nUtZLSNkFZ26Btu8y9qNu3iqqqTP2l77el335b6dqjEoAjj1/PC5O6Ntn5tS5iy6bMPXRt2gRt2gQRgsj82wJ07FzJqg8y/74Dh2zhjRc7AfDm3zox8pNrCxJ1c1E9+pzN0hLls6V4NDA3IuYBSHqIzLOLs/J4zCaxbGFb3p3RgYOO2sTfnupMz49sZ/9DtuxU518vWcm4/zuYC448hE0bSvj2Xe9RkvwJWr64jO9evB9L5rfj8huW0OMjFbRtHyx6tx3LFralV99t/O2pLlRsa5n/UbUEJSXBT//0DvsM2srj9/di9hsduf26fbnpvjls3VLCpg2lfO3MgwCYN6sDx41ew2MT+nDcqLV07FRFp64VrF/beq8+FXP3OZ9nVtfziDuRNFbSdEnTV6yqzGM4jWPzxhJuunwQV964mNLS4KGf9uHib3z42t9rz3Zi/0M289u/z+TnU2Zzx3/0Y+P6zK+7d7/t3DV1Nv/zt1lM+X031qxoQ6eulXz5B4v4ryv35dqzh9BnwDZKinfKuoKrqhJXjR7G50d8lKGHb2TfAzdzzmXLuWHMEC4acRhTHu7B2Bsy//nec3N/DhuxgZ9NmsVHR25gxdKymhZ+a1T9jpZslpao4H/qkofD7wYYfnj7Jn3GMVcV2+Gmywdx8jlr+JfT1zH/nfYse78tX/xEpkWxYmkZV502lPGT/snTv+vOZ69ejgT9Bm/jIwO3sXBuew46clPN/np8pIJBQ7cw45WOHH/GOkaeWs7IU8sBmPSbHpSWNOtfR1HYWN6GN1/qxMdPWsfgYZuY/UZHAP76eHf+89dzAFj9QVtuumJ/ANrvVclxo9ewsbzg/+sUTAAVbinulrqeR2yRIuC2awcyYMhWPnPFCgAGH7yFh9+eyf2vzuL+V2fRq+927pg8m+69K+jVbztvPJ+5DrVmRRsWvduOvgO3smJJGVs3Z/6Crl9bysxpHem//1YA1q5sU1P++L09GXXB6loisT3Vpft2OnauAKBtuyqOOn49789tT8dOlfQbnLkMctTx5Syc0x6Azt0qqJ6T4HNXLePp3/UsTODNSGNMMttc5fPP3TRgiKTBZJLhecAFeTxeXs18tSNT/9CdwQdv5oufGArAJd9awtGnrK+1/oVfXcaPvjqQK04eSgRc9h9L6dKjktf+2oF7btwvcx9/ZG7VGXxw5n/EO2/ox7xZHTLbf21ZTbK0xtW993auvW0BpaWgkuC5J7rx6tSu/OS6ffnOL94lqsSGdaXc9o1BABx2zHouuW4xETDjlb2544aBhT2BQmvBXeNsKPPkTJ52Lp0O/DeZW3ImRMTN9dUffnj7eHXygPqqWDMzamC9s0JZM/NyxWTKq1bvUUbrdlDvOHnCuVnV/eNxd77W0NRhzU1eL4xExCQyD3GbWREp5pZi671abGa7JcdJZlscJ0Uzy0kgKqpa5iBKNpwUzSxnxfyYn5OimeUm3H02M6vha4pmZrtwUjQzSwSi0gMtZmY7eKDFzCwRHmgxM9tZOCmamVUr7gkhnBTNLGduKZqZJSKgsspJ0cyshkefzcwSgbvPZmYpHmgxM9tJHifsLzgnRTPLmbvPZmaJzOizn302M6vh7rOZWYq7z2ZmiUBOimZmaUXce6Z4r5aaWX4ERJWyWhoiaYKk5ZJmpMq+J2mxpDeS5fTUd9+SNFfSbEmnpcpHJWVzJV2fKh8s6ZWk/HeS2jYUk5OimeUsQlktWbgXGFVL+e0RcUSyTAKQNAw4Dzgk2ebnkkollQJ3AKOBYcD5SV2AHyb7OgBYA1zWUEBOimaWs4jslob3E88Bq7M87JnAQxGxNSLmA3OBo5NlbkTMi4htwEPAmZIEnAz8Idn+PuCshg5S5zVFST+lnksHEfGVLE/EzIpIjs8+95Q0PfX57oi4O4vtrpZ0MTAduDYi1gD9gJdTdRYlZQALdykfAfQA1kZERS3161TfQMv0er4zs9YqgOyT4sqIGJ7jEe4EbkqOdBPwY+DSHPex2+pMihFxX/qzpL0iYlP+QzKz5i6fN29HxAfV65LuAZ5IPi4GBqSq9k/KqKN8FdBVUpuktZiuX6cGrylKOkbSLOAfyefDJf28oe3MrFhlN/KczehzrXuX+qY+ng1Uj0xPBM6T1E7SYGAI8CowDRiSjDS3JTMYMzEiAngGODfZfgzwWEPHz+Y+xf8GTksCIiLelHRCFtuZWbFqpJaipAeBE8lce1wEjANOlHREcpQFwBUAETFT0sPALKACuCoiKpP9XA1MBkqBCRExMznEdcBDkv4T+Dvwq4Ziyurm7YhYmBnIqVGZzXZmVoSi8R7zi4jzaymuM3FFxM3AzbWUTwIm1VI+j8zodNaySYoLJR0LhKQy4BrgnVwOYmZFpogfacnmPsUrgavIDGUvAY5IPptZq6Usl5anwZZiRKwELmyCWMyspagqdAD5k83o836SHpe0InlG8TFJ+zVFcGbWDFXfp5jN0gJl033+LfAw0BfYB/g98GA+gzKz5q2xHvNrjrJJintFxK8joiJZfgO0z3dgZtaMRZZLC1Tfs8/dk9Unk6l4HiJzmp+jlqFvM2tFWmjXOBv1DbS8RiYJVp/9FanvAvhWvoIys+ZNLbQVmI36nn0e3JSBmFkLEYLdfISvJcjqiRZJh5KZvLHmWmJE3J+voMysmWuNLcVqksaReTZxGJlriaOBFwAnRbPWqoiTYjajz+cCpwDLIuIS4HCgS16jMrPmrTWOPqdsjogqSRWSOgPL2XnuMjNrTXKbZLbFySYpTpfUFbiHzIj0BuClfAZlZs1bqxx9rhYRX0pW75L0FNA5It7Kb1hm1qy1xqQo6aj6vouI1/MTkpk1d621pfjjer4LMq8ObFT/fGsvTtvniMbereXRvFs+XugQLAdbx7/QODtqjdcUI+KkpgzEzFqIFjyynI2sbt42M9uJk6KZ2Q4q4klmnRTNLHdF3FLMZuZtSfq8pO8mnwdKyuntWGZWPBTZLy1RNo/5/Rw4Bqh+FeF64I68RWRmzV8Rv44gm+7ziIg4StLfASJijaS2eY7LzJqzFtoKzEY2SXG7pFKSX4OkXhT1u7zMrCEttWucjWyS4njgUaC3pJvJzJrznbxGZWbNV7Ty0eeIeEDSa2SmDxNwVkS8k/fIzKz5as0tRUkDgU3A4+myiHg/n4GZWTPWmpMi8Cd2vMCqPTAYmA0ckse4zKwZa9XXFCPio+nPyew5X6qjuplZi5bzEy0R8bqkEfkIxsxaiNbcUpT09dTHEuAoYEneIjKz5q21jz4DnVLrFWSuMT6Sn3DMrEVorS3F5KbtThHx700Uj5k1c6K4B1rqfPZZUpuIqASOa8J4zKwlaKRXnEqaIGm5pBmpsu6Spkiak/zslpRL0nhJcyW9lX5liqQxSf05ksakyj8m6e1km/GSGnwgu74JIV5Nfr4haaKkiySdU700fLpmVpQad5ace4FRu5RdD0yNiCHA1OQzwGhgSLKMBe6ETBIFxgEjgKOBcdWJNKnzhdR2ux7rQ7KZJac9sIrMO1nOAD6d/DSz1qoqy6UBEfEcsHqX4jOB+5L1+4CzUuX3R8bLQFdJfYHTgCkRsToi1gBTgFHJd50j4uWICOD+1L7qVN81xd7JyPMMdty8XXMuDe3YzIpXnq8p9omIpcn6MqBPst4PWJiqtygpq698US3l9aovKZYCe7NzMqzmpGjWmmWfAXpKmp76fHdE3J31YSJCatphnfqS4tKIuLHJIjGzliG3t/mtjIjhOR7hA0l9I2Jp0gVenpQvBgak6vVPyhYDJ+5S/mxS3r+W+vWq75piy5w218zyLs+vI5gIVI8gjwEeS5VfnIxCjwTWJd3sycCpkrolAyynApOT78oljUxGnS9O7atO9bUUT9m98zGzotdIHVpJD5Jp5fWUtIjMKPItwMOSLgPeAz6bVJ8EnA7MJTNz1yUAEbFa0k3AtKTejRFRPXjzJTIj3B2AJ5OlXnUmxdROzcx20liP+UXE+XV89aFGWTKCfFUd+5kATKilfDpwaC4x+RWnZpab3K4ptjhOimaWE1HcAw5OimaWO7cUzcx2KOYJIZwUzSx3TopmZglPMmtmtgu3FM3MdvA1RTOzNCdFM7Md3FI0M6sWZDWBbEvlpGhmOSn2F1c5KZpZ7pwUzcx2UBRvVnRSNLPceJYcM7Od+ZqimVmKH/MzM0tzS9HMLLFnL6Vq9pwUzSx3TopmZhm+edvMbBeqKt6s6KRoZrnxfYqWi/temcXmDaVUVUFlhfjy6AP5/LXLGH3BKtatzvy6/+cHfZn2l84FjrT4/eDYZzip33us2tKBTz3+OQBG7fsuXzl8Ovt3WcNnJp3DjFW9AfjXwf/k8kPerNl2aLdVnPXEubyzpienD5rLFz/6OqUKnlm0L7e+PhKA8w+cyYVDZ1IVYmNFGTe8dAJz13Vv+hMtAN+SsxskTQDOAJZHRE4vo27pvvlv+1O+eudf7aP39OIPd/UuUESt0x/nDuXX/ziUW4/7S03ZnLXduerZ07hp5F93qjtx/oFMnH8gAAd2XcWdJ03mnTU96dpuC9d97GXOfuIzrN7agR8e9xeO+cgiXlrWn8fnD+HBfx4CwMn9F/Ct4S9x2dRPNd0JFlIRtxRL8rjve4FRedy/Wb2mLd+HdVvb7VT27rpuzC/vWu92ZwyeyxPz9wdgwN7lLCjvwuqtHQD429L+nLbvPAA2bG9bs81ebbYXc574EEV2S0uUt5ZiRDwnaVC+9t9shfivB+dBwJ9+3YMnH+gBwKcvWckp565hzlsduPv7+7Bhna9cNFefGvQuVz6T+Xv+3vou7Nd5Lf06lrNs0958csB8ykp29B0vHDqDS4e9RVlJJRc9/elChdy0AvCEEPkjaSwwFqA9exU4mj339bMOYNWyMrr02M4tD81j4dx2PHFfD357ex8iYMw3lzF23BJu+/rAQodqtTi85wdsrmjDnLWZa4Pl29ox7pXj+ckJf6YK8fryPgzsVF5T/4HZh/LA7EP59OA5fOmw17nuxZMLFXqTKuZrivnsPmclIu6OiOERMbyMdg1v0MytWlYGwLpVZbz4VBcOOnITa1eWUVUlIsSTD/Rg6BGbCxyl1eVTg+byxPwDdir7y6JBnPvkOXz2ybOZX96VBeVdPrTdE/MP4JMDFjRRlIVVfZ9isXafC54Ui0m7DpV06FhZs/6x/7OeBf9oT/fe22vqHDt6HQtmty9UiFYPEYwe9C5/WrBzUuzePvNHrHPbrVw4dCYPzzkYgH07ra2pc1L/92pNlkUpIvulBSp497mYdOtVwbhfLQCgtE3wzKPdmP5sZ74x/n32P2QzEfDBoraM/2b/wgbaStx+/J85us8SurXfwvOf+TU/eXM467a257tHv0D39pu55+QneWdNDy798xkAfLzPEpZt3JuFG3a+XeqGj7/IQd1WAfCztz7GgvVdAbjooBkc23cxFVUlrNvWjm++eFKTnl8htdRWYDYUecrmkh4ETgR6Ah8A4yLiV/Vt01ndY4ROyUs8lh/zbjmm0CFYDhaNv52tixZqT/bRqWv/OPKEa7Kq+/zj33wtIobvyfGaWj5Hn8/P177NrLCKuaXo7rOZ5SaAyuLNik6KZpazYm4pevTZzHLXSKPPkhZIelvSG5KmJ2XdJU2RNCf52S0pl6TxkuZKekvSUan9jEnqz5E0Zk9OzUnRzHLWyPcpnhQRR6QGZK4HpkbEEGBq8hlgNDAkWcYCd0ImiQLjgBHA0cC46kS6O5wUzSw3kcOye84E7kvW7wPOSpXfHxkvA10l9QVOA6ZExOqIWANMYQ/mXXBSNLOcCFBlZLUAPSVNTy1jd9ldAE9Lei31XZ+IWJqsLwP6JOv9gIWpbRclZXWV7xYPtJhZzpT9/c0rG7hP8V8iYrGk3sAUSf9IfxkRITXtsI5bimaWm0bsPkfE4uTncuBRMtcEP0i6xSQ/lyfVFwMDUpv3T8rqKt8tTopmlqPGefZZUkdJnarXgVOBGcBEoHoEeQzwWLI+Ebg4GYUeCaxLutmTgVMldUsGWE5NynaLu89mlrNG6tD2AR6VBJlc9NuIeErSNOBhSZcB7wGfTepPAk4H5gKbgEsAImK1pJuAaUm9GyNi9e4G5aRoZrlrhDkTImIecHgt5auAD02CEJmJGq6qY18TgAl7HBROimaWq6B6ZLkoOSmaWe6KNyc6KZpZ7nK4JafFcVI0s9w5KZqZJQIo4hdXOSmaWU5EuPtsZraTquJtKjopmllu3H02M9uZu89mZmlOimZm1Vrui+6z4aRoZrnx2/zMzHbma4pmZmlOimZmiQCqnBTNzBIeaDEz25mToplZIoDK4n2kxUnRzHIUEE6KZmY7uPtsZpbw6LOZ2S7cUjQzS3FSNDNLREBlZaGjyBsnRTPLnVuKZmYpTopmZtXCo89mZjUCwjdvm5ml+DE/M7NEhF9xama2Ew+0mJntEG4pmplV8ySzZmY7eEIIM7MdAgg/5mdmlghPMmtmtpNw99nMLKWIW4qKZjSKJGkF8F6h48iDnsDKQgdhOSnWf7N9I6LXnuxA0lNkfj/ZWBkRo/bkeE2tWSXFYiVpekQML3Qclj3/m7VeJYUOwMysOXFSNDNLcVJsGncXOgDLmf/NWilfUzQzS3FL0cwsxUnRzCzFSTGPJI2SNFvSXEnXFzoea5ikCZKWS5pR6FisMJwU80RSKXAHMBoYBpwvaVhho7Is3Au0qJuNrXE5KebP0cDciJgXEduAh4AzCxyTNSAingNWFzoOKxwnxfzpByxMfV6UlJlZM+akaGaW4qSYP4uBAanP/ZMyM2vGnBTzZxowRNJgSW2B84CJBY7JzBrgpJgnEVEBXA1MBt4BHo6ImYWNyhoi6UHgJWCopEWSLit0TNa0/JifmVmKW4pmZilOimZmKU6KZmYpTopmZilOimZmKU6KLYikSklvSJoh6feS9tqDfd0r6dxk/Zf1TVYh6URJx+7GMRZI+tBb3+oq36XOhhyP9T1J/55rjGa7clJsWTZHxBERcSiwDbgy/aWk3XqPd0RcHhGz6qlyIpBzUjRriZwUW67ngQOSVtzzkiYCsySVSrpV0jRJb0m6AkAZP0vmd/wz0Lt6R5KelTQ8WR8l6XVJb0qaKmkQmeT7taSVerykXpIeSY4xTdJxybY9JD0taaakXwJq6CQk/X9JryXbjN3lu9uT8qmSeiVl+0t6KtnmeUkHNcpv0yyxWy0LK6ykRTgaeCopOgo4NCLmJ4llXUR8XFI74EVJTwNHAkPJzO3YB5gFTNhlv72Ae4ATkn11j4jVku4CNkTEj5J6vwVuj4gXJA0k89TOwcA44IWIuFHSp4Bsnga5NDlGB2CapEciYhXQEZgeEV+T9N1k31eTeaHUlRExR9II4OfAybvxazSrlZNiy9JB0hvJ+vPAr8h0a1+NiPlJ+anAYdXXC4EuwBDgBODBiKgElkj6Sy37Hwk8V72viKhrXsFPAMOkmoZgZ0l7J8c4J9n2T5LWZHFOX5F0drI+IIl1FVAF/C4p/w3wx+QYxwK/Tx27XRbHMMuak2LLsjkijkgXJMlhY7oI+HJETN6l3umNGEcJMDIittQSS9YknUgmwR4TEZskPQu0r6N6JMddu+vvwKwx+Zpi8ZkMfFFSGYCkAyV1BJ4DPpdcc+wLnFTLti8DJ0ganGzbPSlfD3RK1Xsa+HL1B0lHJKvPARckZaOBbg3E2gVYkyTEg8i0VKuVANWt3QvIdMvLgfmS/i05hiQd3sAxzHLipFh8fknmeuHrycuXfkGmR/AoMCf57n4yM8HsJCJWAGPJdFXfZEf39XHg7OqBFuArwPBkIGcWO0bBv08mqc4k041+v4FYnwLaSHoHuIVMUq62ETg6OYeTgRuT8guBy5L4ZuJXPFgj8yw5ZmYpbimamaU4KZqZpTgpmpmlOCmamaU4KZqZpTgpmpmlOCmamaX8L1QAJ2IeIKaHAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualization(dtModel)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f51e816b",
   "metadata": {},
   "source": [
    "### Linear Support Vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1bb06bff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mislabeled out of a total of 37076 test entries: 547\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of mislabeled out of a total of %d test entries: %d\" %  (urls_test.shape[0], \n",
    "                                                                       (ifSpam_test != lsvcPredict).sum()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "c27185a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "successRate = 100.0 * f1_score(ifSpam_test, lsvcPredict, average='micro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "9550e491",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Success Rate was calculated as % : 98.52465206602655 with LinearSVC\n"
     ]
    }
   ],
   "source": [
    "print(\"The Success Rate was calculated as % : \" + str(successRate) + \" with LinearSVC\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "3a7713ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualization(lsvcModel)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bb7d148",
   "metadata": {},
   "source": [
    "### Stochastic Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "9cd7a13d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mislabeled out of a total of 37076 test entries: 1752\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of mislabeled out of a total of %d test entries: %d\" %  (urls_test.shape[0], \n",
    "                                                                       (ifSpam_test != sgdcPredict).sum()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "e6805bb0",
   "metadata": {},
   "outputs": [],
   "source": [
    "successRate = 100.0 * f1_score(ifSpam_test, sgdcPredict, average='micro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "6545b107",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Success Rate was calculated as % : 95.27457115114899 with Stochastic Gradient Descent\n"
     ]
    }
   ],
   "source": [
    "print(\"The Success Rate was calculated as % : \" + str(successRate) + \" with Stochastic Gradient Descent\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "df2cda83",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAve0lEQVR4nO3deZwU1bn/8c+3ewZmgGEREGUdZFFBFnFcUdxu3JJINCoak7gbEzFGo4n3xgW3X9xz4xIRjaLeiBo1BhIj0YBbCOqg4wKIIqCAIIiyL7M9vz+qemiGYbqBqe6Z6ef9evVruqpO1XlqGPrpc6rqHJkZzjnnclcs2wE455zLLk8EzjmX4zwROOdcjvNE4JxzOc4TgXPO5bi8bAewvTp16mTFxcXZDsM555qUGTNmfGVmneva1uQSQXFxMaWlpdkOwznnmhRJn21rm3cNOedcjvNE4JxzOc4TgXPO5ThPBM45l+M8ETjnXI6LLBFIeljSMkkfbmO7JN0taa6k9yUNiyoW55xz2xZli2A8cFw9248H+oWvC4H7I4zFOefcNkT2HIGZvSapuJ4iI4HHLBgHe7qk9pJ2N7MlUcTz9oKvef3j5du3k8T3hnZlj85tqKiqpryymsoqo7yqmorwVW1QVV1NVTVUVVvwsuBny7wYA7u2xQwqq43qcH1ltVEdluvYugWSojhl55xLSzYfKOsGLExaXhSu2yoRSLqQoNVAz549d6iydz77hnumzt2ufczg7n99skP1bY92hfms2lDBAb13qUkUVdVB0vl6XTnL1mzihwf1RIjPv17P+Yf1pmv7QszAzKg2MIzqaqg2C9YTrK9ZTpQLf+61WxFFBXnEJCQ8GW2HxO+wKkzudZdJeo/VuT7YtuVx615f++Cpj72t49Z3bGMbB0hzn23VXzuGbU2BsvXvZvOKymqjbUE++XERj4n8eIz8eIx4zP9uG0KTeLLYzMYB4wBKSkp2aCadnxzeh58c3me79jn7kbdYumojR+61K6s2VNBrl1bkxWO0iIu8eIy1GyvpXNSSWEzEFfyBBq/gg3Xa3K8ozI8Tj8WIxyAWE3kxEQvL3jF5DoUt4nRrX0hePEZM0CI/TjwsF4+Jj5auAeD/pn9eE9er29uySYMEcYnK6uDXW9yxFQqTRExCUJM0Ej8TLaBeHVvXLFdUVQfJrMpYunojlVXVtMyPA1smrSA5UbPeoCaBmUEQRiKp1SoTvi+vrGZTZTV77VZUk8gUngvJPwneLF65gd3aFtS0zKpsc8usOtGiS1pXFbbcqo2a9dXhObjG54RBu3Heob3Zr9cu2Q6lyVGUM5SFXUN/M7N96tj2APCKmU0Il+cAR6TqGiopKbFcHWJiY0UV90z5hC9Xb+Kwfp2QREzBB10s/Faf+KCOhT9JWk6Um7t8LQu+Wk+HVvk1rYbE641PvuK9RasYObRrTQvCkloWiQ/CajPeW7iS3dsX1LREEt/Q8uMiLxYjLy6+WltOcccggQbJJIwTIIwpDDP8KWKxYGPt9ZuXg/fvLVxJpzYta743Wpg8Nr8naVvQpbd2YyVd2hYQj2mLBB4kZzYnc23eHgu3x0RN2cTyinXl7Nq2Zc2/USLpwOZEFJ5qnevr22eLMrU2bOt4W67f9nG3WEwutwNxbjuWbVe6zTi33AMJvl5Xzhtzv+LovXaloipI0Gs3VfJ/0z9jxbpytmV4346M6NeZucvWctYhxezTrd02y+YCSTPMrKTObVlMBN8GRgMnAAcCd5vZAamOmcuJwDm3bR8uXsX9r3zKa58sZ83GyjrLDO/bkUP6dOKHB/WiXWF+hiPMrqwkAkkTgCOATsCXwHVAPoCZjVXwNeBegjuL1gPnmFnKT3hPBM65dFRUVbOhoop3PvuGiWVf8Ny7i7fYvl+vDtz0vX3Ye/e2WYows7LWIoiCJwLn3I6orjbWhF1Kt0+eU7P+8fMO4LB+dY7O3Kx4InDOuSTV1cZj/1nAmEmzatbd94NhfHvw7lmMKlr1JYImcdeQc841pFhMnD28Nx3btOSfs75k0ntfcPET77Bu02BO279HtsPLOB9ryDmXs747pCv3nLEvvz99KAC/evZ9rnm+zlFxmjVPBM65nDdyaDfe+s3RADw+/TOeLl1IRVV1lqPKHE8EzjkH7FpUwJXH7gnAr555n36/+Qe/fPo9NlZUZTmy6HkicM650MVH9mXm9cdyyn7dyY+LZ99ZxF7XvNjsu4s8ETjnXJLWLfO449QhzLnxePbarQgIuosOuPllVm+syHJ00fBE4JxzdYjFxIu/GME/LxsBwLI1mxg85p/c9dLHWY6s4XkicM65evTvUsTsG47jxpEDgWBE4uKr/s6qDc2ndeCJwDnnUihsEedHBxfzj0sPq1k35Pp/8pPHS1m3qe5xjZoSTwTOOZemvXdvy4Jbvs1vTx4EwOSZXzLwusmMfuKdLEe2czwROOfcdjrjgJ58+v9OYMx3BwDwt/eX8O7n32Q5qh3nicA553ZAPBym4tmfHgLASX+YxkuzvsxyVDvGE4Fzzu2EYT3bc8SeweilFzxWysj7/k11E5vGzhOBc87tBEmMP+cAyq79FhDMnDf81iks/Hp9liNLnycC55xrAO1btWD+b08AYMmqjRx221QWfLUuy1GlxxOBc841EEm8P+YYTh7WDYAj7niFqibQTeSJwDnnGlDbgnzuPHUIu7crAGDsq59mOaLUPBE451wDk8TLlx8OwO2T51B81d9ZtmZjlqPaNk8EzjkXgdYt8/jxwb1qlg+4+V9ZjKZ+ngiccy4iN4zchwW3fJvuHQoB+OfMpVmOqG6RJgJJx0maI2mupKvq2N5L0r8kvS/pFUndo4zHOeey4fHzDgTgwsdnUNkIZz6LLBFIigP3AccDA4AzJA2oVewO4DEzGwzcAPw2qniccy5bendqzfC+HQHof/U/uG/qXKZ+tCzLUW0WZYvgAGCumc0zs3LgSWBkrTIDgCnh+6l1bHfOuWbh/847kO8O6Uq1BReQzxn/Ni83kiEpokwE3YCFScuLwnXJ3gNODt+fBBRJ6lj7QJIulFQqqXT58uWRBOucc1GSxD1n7MuCW77N8xcPB+D8x0r5au2mLEeW/YvFVwCHS3oXOBxYDGw1U7SZjTOzEjMr6dy5c6ZjdM65BjW0R/uah86G3zIlRenoRZkIFgM9kpa7h+tqmNkXZnayme0L/CZctzLCmJxzrlG467ShHNB7FzZVVvP2gq+zGkuUieBtoJ+k3pJaAKcDE5MLSOokKRHDfwMPRxiPc841Kr85YW8ATh37H6bPW5G1ONJKBJJ2lXSSpIslnSvpgKQP8DqZWSUwGpgMzAaeNrOZkm6QdGJY7AhgjqSPgS7AzTt8Js4518QM6dGeQd3aAXD6uOlsqtyqZzwjZLbtAZEkHQlcBewCvAssAwqA/kAf4BngTjNbHX2ogZKSEistLc1Udc45F7mxr37KLf/4iNtPGcypJT1S77ADJM0ws5K6tqVqEZwAXGBm+5vZhWZ2tZldYWYnAkMIksO3Gjhe55zLKWcfUgyQtRnO8urbaGZX1rOtEni+oQNyzrlcU5Afp1WLOP+c9SXryytp1aLej+YGt8MXiyWd05CBOOdcLjv/0N4APPja/IzXvTN3DV3fYFE451yOu2DEHgAUtsj84131tj8kvb+tTQR3+TjnnGsA+fEgAVRmYUazVB1RXYBjgW9qrRcwLZKInHMuB+XFBEBVVeNLBH8D2phZWe0Nkl6JIiDnnMtF8TARNLoWgZmdV8+2HzR8OM45l5skEY8pK5PdZ3vQOeecc0k+/3p9xuv0ROCcc41EVbUxJQsT1ngicM65RmTtpsqM1+mJwDnnGolzhhdTkJ/5j+W0a5Q0rr5l55xzOyfxLEGmbU+tD6RYds45txPiMVFdnfl6004EZjajvmXnnHM7Jy5RmYVMkGqIiUnANm9qDYejds451wBiMVFtYGZIyli9qZ4sviMjUTjnnNs8zES1kRdvJInAzF5NvJdUCPQ0szmRR+WcczkoMczEuvIq2hVm7sJxunMWfxcoA14Ml4dKmljvTs4557ZLokXwzue1x/mMVropZwxwALASIByErnckETnnXI46Ys9dAShd8HVG6003EVSY2apa6zI/MpJzzjVje+5WBMAnX67NaL3pJoKZkn4AxCX1k3QPacxHIOk4SXMkzZV0VR3be0qaKuldSe9LOmE743fOuWYn0wOQppsILgEGApuACcBq4Bf17SApDtwHHA8MAM6QNKBWsauBp81sX+B04A9pR+6cc83QwK5t+WLlhozWmer2UQDMbD3wG0m3Bou2Jo3dDgDmmtk8AElPAiOBWcmHBtqG79sBX6QbuHPONUd77lbEc+8szmid6d41tL+kD4D3gQ8kvSdpvxS7dQMWJi0vCtclGwP8UNIi4AWClkdd9V8oqVRS6fLly9MJ2TnnmqSWeXEA1mVwFNJ0u4b+CPzMzIrNrBi4GHikAeo/AxhvZt2BE4DHJW0Vk5mNM7MSMyvp3LlzA1TrnHONU+9OrYDM3kKabiKoMrPXEwtm9gaQKl0tBnokLXcP1yU7D3g6POZ/gAKgU5oxOedcs3PwHsFH4Pyv1mWsznoTgaRhkoYBr0p6QNIRkg6X9AfglRTHfhvoJ6m3pBYEF4NrP4T2OXB0WNfeBInA+36cczlrn27BZdO3F2SuRZDqYvGdtZavS3pf7w1OZlYpaTQwGYgDD5vZTEk3AKVmNhH4JfCgpMvC451tZv58gnMuZyUGm9tYUZWxOlONNXTkzhzczF4guAicvO7apPezgOE7U4dzzjU3A3ZvS3ll5oajTuv2UQBJ3yZ4lqAgsc7MbogiKOecy2V5cbFgRSO5RpAgaSwwiuD2TgGnAr0ijMs553JW9w6FZG4Q6vTvGjrEzH4MfGNm1wMHA/2jC8s553JXqxZ5LFixPmP1pZsIEs87r5fUFagAdo8mJOecy22tWgQPlW2qzMwF43QTwd8ktQduB94BFhCMOeScc66B7dK6BQCr1ldkpL50xxq6MXz7rKS/AQV1DEvtnHOuAfTqGDxdvGTVRnZtW5Ci9M5LNXn9yfVsw8yea/iQnHMut3Vs3RKAT5evZUiP9pHXl6pF8N16thngicA55xrYrm1bZrS+VA+UnZOpQJxzzgWKCvIBqKjKzENl6V4sds45lyEt4sFH88wvVmekPk8EzjnXyHRoFbQIlq3elJH6PBE451wjkxePUdQyjzYFaY8CtFPSHWKilaRrJD0YLveT9J1oQ3POudzVoXULqjI0i326LYJHCCauPzhcXgzcFElEzjnnyIuJykaWCPqY2W0EQ0skJrPP5JhIzjmXU+IxUdnI7hoql1RIOBmNpD4ELQTnnHMRiMdERVVmWgTpXokYA7wI9JD0J4LJZM6OKCbnnMt5ZjBv+dqM1JXuWEP/lDQDOIigS+hSM/sq0siccy6HbaqsYrd20Y8zBGkmAkmTgCeAiWaWuWlznHMuR3VtX9joniy+AzgMmCXpGUmnSMpMqnLOuRwUb2x3DZnZq2b2M2AP4AHgNGBZqv0kHSdpjqS5kq6qY/vvJJWFr48lrdzO+J1zrlnKiyljzxFsz+T1hQSjkY4ChgGPpigfB+4DvgUsAt6WNNHMZiXKmNllSeUvAfbdruidc66ZisdiVGborqF0nyx+GpgNHAXcS/BcwSUpdjsAmGtm88ysHHgSGFlP+TPwWc+ccw4IWgQLv87MvMXptgj+CJxhZtszgWY3YGHS8iLgwLoKSuoF9AambMfxnXOu2Vq2ZiP5eZkZDi7VDGVHmdkUoDUwUtryYeIGnKHsdOCZbSUaSRcCFwL07Nmzgap0zrnGq3+XIhZ9syEjdaVqERxO8C29rpnKUs1QthjokbTcPVxXl9OBi7d1IDMbB4wDKCkpyUynmXPOZZEUDuWQAalmKLsufHuDmc1P3iapd4pjvw30C8stJviw/0HtQpL2AjoA/0k3aOeca+6CeeEzU1e6HVDP1rHumfp2MLNKYDQwmeBC89NmNlPSDZJOTCp6OvCkWaZO2TnnGr+YIFMfi6muEewFDATaSTo5aVNbIOUDZWb2AvBCrXXX1loek26wzjmXK2IS1Y0hEQB7At8B2rPldYI1wAURxeScczkvSASZqSvVNYK/An+VdLCZeR++c85liETjaBFI+lU4Ic0PJJ1Re7uZ/TyyyJxzLoeJzF0sTtU1NDv8WRp1IM455zZrNBeLzWxS+LNmXCFJMaCNma2OODbnnMtZsZhYV749gznsRF3pFJL0hKS2kloDHxIMR31ltKE551zuWrG2HMhMqyDd5wgGhC2A7wH/IBgX6EdRBeWcc7mud6dWABmZkyDdRJAvKZ8gEUw0swoy9/Szc87lnBbhgHPlldHPUpZuIngAWEAw+Nxr4Wihfo3AOeci0iIefDxvaiyJwMzuNrNuZnaCBT4Djow4Nuecy1mJ0Z7nLF0TeV3pXixuJ+kuSaXh606C1oFzzrkI7LlbERDMSxC1dLuGHiYYVuK08LUaeCSqoJxzLtf12CW4WPzJl2sjryvdGcr6mNn3k5avl1QWQTzOOeeA3doG43pWVDWSawTABkmHJhYkDQcyM3WOc87loHhMtG4RpyoDt4+m2yK4CHhMUrtw+RvgrGhCcs45B8HTxZl4jiBlIpA0FOhLMIHMYgAfXsI556KXF8vMnAT1dg1JuhZ4Gvg+8HdglCcB55zLDEl8sTL6u4ZStQhGAUPNbL2kjsCLwIORR+Wcc44N5VV8/vW6yOtJdbF4k5mtBzCzFWmUd84510B2a1fAuk3Rj0CaqkWwh6SJ4XsBfZKWMbMT697NOefczhqwe1s+/GJV5PWkSgQjay3fEVUgzjnntlSQH6eyKst3DZnZqztzcEnHAb8H4sBDZnZLHWVOA8YQjGb6npn9YGfqdM655iIvpow8R5DqrqFJkr4bDkFde9sekm6QdO429o0D9wHHAwOAMyQNqFWmH/DfwHAzGwj8YsdOwznnmp9YDJauzv5dQxcAlwP/K+lrYDlQABQDnwL3mtlft7HvAcBcM5sHIOlJgq6mWbWOf5+ZfQNgZst28Dycc67ZSbQGqqqNeEyR1ZOqa2gp8CvgV5KKgd0Jhpb4OHE3UT26AQuTlhcBB9Yq0x9A0r8Juo/GmNmLtQ8k6ULgQoCePXumqNY555qH3p3aAMHkNIUt4pHVk+4QE5jZAoLJaRq6/n7AEUB3gklvBpnZylp1jwPGAZSUlPjMaM65nJA8S1mUiSDK5wIWAz2SlruH65ItIpz60szmAx8TJAbnnMt5VdXByKNrNlVEWk+UieBtoJ+k3pJaEIxVNLFWmecJWgNI6kTQVTQvwpicc67JaN+qBQAbK6IdijqyRGBmlcBoYDIwG3jazGaGdxolHkSbDKyQNAuYClwZPsHsnHM5r11hcMPmmo3RtgjSukYQzj8wBugV7iPAzGyP+vYzsxeAF2qtuzbpvRHclXT5dkXtnHM5oCA/uC4wZ+ka9u3ZIbJ60r1Y/EfgMmAGEP3AF8455+i3a5uM1JNuIlhlZv+INBLnnHNbaBXeKfRxxPMWp5sIpkq6HXgO2JRYaWbvRBKVc8452rQMPqLbt9pqcIcGlW4iSDwIVpK0zoCjGjYc55xzCYmniaMebyitRGBmR0YahXPOua1IIiYin64yrdtHJbWTdJek0vB1Z9JE9s455yISz8AE9uk+R/AwsAY4LXytBh6JKijnnHOBmER1Y+gaAvqY2feTlq+XVBZBPM4555IYsHzNppTldka6LYINkg5NLIQPmG2IJiTnnHMJ5ZXVrN5YGWkd6bYIfgo8Gl4XEPA1cHZUQTnnnAt0bVfAZyvWRVpHuncNlQFDJLUNl1dHGZRzzrlAm4I82he2iLSOehOBpB+a2f9JurzWegDM7K4IY3POuZy3W7tCVm/I7qBzrcOfRZFG4Zxzrk4t4qKiKtphqFNNVflA+PP6SKNwzjlXp/x4jNlLou2NT/eBstsktZWUL+lfkpZL+mGkkTnnnGPVhoqaCWqiku7to8eEF4i/QzBvcV/gyqiCcs45F+jfpSjyrqF0E0GiC+nbwJ/NbFVE8TjnnEvSMj/GmoifI0g3EfxN0kfAfsC/JHUGNkYXlnPOOaAmCUQ5XWVaicDMrgIOAUrMrAJYB4yMLCrnnHMA9A9nKSuvjK57KNVzBEeZ2RRJJyetSy7yXFSBOeec2zwnQZTjzqV6juBwYArw3Tq2GZ4InHMuUrFYzQO8kdWR6jmC68Kf5+zIwSUdB/weiAMPmdkttbafDdwOLA5X3WtmD+1IXc451xzFwl6YqggTQbrPEfw/Se2TljtIuinFPnHgPuB4YABwhqQBdRR9ysyGhi9PAs45lySu6LuG0r1r6HgzW5lYMLNvgBNS7HMAMNfM5plZOfAkfoHZOee2S+KybJST06SbCOKSWiYWJBUCLespD9ANWJi0vChcV9v3Jb0v6RlJPeo6kKQLE9NkLl++PM2QnXOu6dt8sTj7ieBPBM8PnCfpPOAl4NEGqH8SUGxmg+s7ppmNM7MSMyvp3LlzA1TrnHNNQ2VVkABWRTgCabrzEdwq6T3gv8JVN5rZ5BS7LQaSv+F3Z/NF4cRxVyQtPgTclk48zjmXK9oUBB/T2bx9NNlsoNLMXpbUSlKRma2pp/zbQD9JvQkSwOnAD5ILSNrdzJaEiyeGdTjnnAu1bhl8TFdVZ+mBsgRJFwAXArsAfQj6+scCR29rHzOrlDQamExw++jDZjZT0g1AqZlNBH4u6USgEp/+0jnntpIXXiNIdBFFUkea5S4muAvoTQAz+0TSrql2MrMXgBdqrbs26f1/A/+ddrTOOZdjEheLs/4cAbApvAUUAEl5BE8WO+eci1CiRVDVCG4ffVXS/wCFkr4F/Jngjh/nnHMRSrQIohx0Lt1E8GtgOfAB8BOC7p6rowrKOedcIJEIvli5IbI6Ul4jCIeKmGlmewEPRhaJc865rXRpWwDAinXlKUruuJQtAjOrAuZI6hlZFM455+rUvlU+EO0QE+neNdQBmCnpLYJJaQAwsxMjico55xwALfPiSPDN+iw/WQxcE1kEzjnn6iXgrflfR3b8VDOUFQAXAX0JLhT/0cyinUXZOefcFvLjMdoWbs9AENsn1TWCR4ESgiRwPHBnZJE455yr07CeHRBKXXAHpUoxA8xsEICkPwJvRRaJc865OuXFxcbKqsiOn6pFUHN1wruEnHMuO/JiivTJ4lQtgiGSVofvRfBk8erwvZlZ28gic845BwTzFs9esjp1wR2UavL6eGQ1O+ecS8uajZV0bpNqUsgdl+4QE84557KkV8dWkR7fE4FzzjVyMalRDEPtnHMuS2IxRTpVpScC55xr5GKKdqwhTwTOOdfIxWPeNeScczktJrGhPHsPlDnnnMuyTZVVbKqsjuyhMk8EzjnXyLUrbAFARVU001VGmggkHSdpjqS5kq6qp9z3JZmkkijjcc65pqhDODlNk2sRhFNc3kcwaukA4AxJA+ooVwRcCrwZVSzOOdeUJeYtjuqCcZQtggOAuWY2z8zKgSeBkXWUuxG4FdgYYSzOOddkbaoMuoTWboxm7M8oE0E3YGHS8qJwXQ1Jw4AeZvb3+g4k6UJJpZJKly9f3vCROudcI9auMOgaapLXCOojKQbcBfwyVVkzG2dmJWZW0rlz5+iDc865RqSoIBgftMldIwAWAz2SlruH6xKKgH2AVyQtAA4CJvoFY+ec21JM4TWCJpgI3gb6SeotqQVwOjAxsdHMVplZJzMrNrNiYDpwopmVRhiTc841OXlN9WJxOKPZaGAyMBt42sxmSrpB0olR1eucc81NLEwElVXRJIJUM5TtFDN7AXih1rprt1H2iChjcc65pirRIqiOqEUQaSLIlIqKChYtWsTGjX4HqsuugoICunfvTn5+frZDcc1IokWwJqLbR5tFIli0aBFFRUUUFxej8KKKc5lmZqxYsYJFixbRu3fvbIfjmpHEENSbKqMZeK5ZjDW0ceNGOnbs6EnAZZUkOnbs6C1T1+B2aR2MNSSi+YxrFokA8CTgGgX/O3RRiEd8jaDZJALnnGuuEs8RRDVJmSeCBnLzzTczcOBABg8ezNChQ3nzzWAMvcrKSv7nf/6Hfv36MXToUIYOHcrNN99cs188Hmfo0KEMHDiQIUOGcOedd1Jdvfkx8rfeeosRI0aw5557su+++3L++eezfv16xo8fz+jRoxss/hNOOIGVK1cCcPfdd7P33ntz5plnMnHiRG655ZadOvaSJUv4zne+s8W6X/ziF3Tr1m2Lcx0/fjydO3dm6NChDBgwgAcffHCn6gWYP38+Bx54IH379mXUqFGUl5dvVaa8vJxzzjmHQYMGMWTIEF555ZWabRMmTGDQoEEMHjyY4447jq+++gqAK664gilTpux0fM6lI9HQjOqBMsysSb32228/q23WrFlbrcukadOm2UEHHWQbN240M7Ply5fb4sWLzczs17/+tZ111lm2YcMGMzNbvXq1XXfddTX7tm7duub9l19+aUcffbRde+21Zma2dOlS69mzp02bNq2mzJ///GdbunSpPfLII3bxxRdHcj577rmnLVy4cIf2raio2GrdFVdcYc8//3zNclVVlfXs2dMOPPBAmzJlSs365HP68ssvrVOnTrZ06dIdiiPh1FNPtQkTJpiZ2U9+8hP7wx/+sFWZe++9184+++yaeocNG2ZVVVVWUVFhnTt3tuXLl5uZ2ZVXXlnzb7dgwQL71re+VWed2f57dM3P7CWrrNev/2YvvP/FDh8DKLVtfK42i7uGkl0/aSazvljdoMcc0LUt13134Da3L1myhE6dOtGyZUsAOnXqBMD69et58MEHWbBgAQUFBQAUFRUxZsyYOo+z6667Mm7cOPbff3/GjBnDfffdx1lnncXBBx9cU+aUU07Zar9JkyZx0003UV5eTseOHfnTn/5Ely5dePXVV7n00kuBoO/6tddeY+3atYwaNYrVq1dTWVnJ/fffz2GHHUZxcTGlpaVcffXVzJs3j+OPP55zzz2XDh06UFpayr333svy5cu56KKL+PzzzwH43//9X4YPH86YMWP49NNPmTdvHj179mTChAlbxPfss89y00031Sy/8sorDBw4kFGjRjFhwgSOPPLIOn8Xffr04bPPPqNLly7b/N3Xx8yYMmUKTzzxBABnnXUWY8aM4ac//ekW5WbNmsVRRx1VU2/79u0pLS1l3333xcxYt24dHTt2ZPXq1fTt2xeAXr16sWLFCpYuXcpuu+22Q/E5ly7vGmoCjjnmGBYuXEj//v352c9+xquvvgrA3Llz6dmzJ0VFRWkfa4899qCqqoply5bx4Ycfst9++6Xc59BDD2X69Om8++67nH766dx2220A3HHHHdx3332UlZXx+uuvU1hYyBNPPMGxxx5LWVkZ7733HkOHDt3iWGPHjqVr165MnTqVyy67bIttl156KZdddhlvv/02zz77LOeff37NtlmzZvHyyy9vlQTmz59Phw4dapIkBN0tZ5xxBieddBJ///vfqaio2Oqc5s2bx7x582o+eBPmzJlT08VW+5Xo2kpYsWIF7du3Jy8v+L7TvXt3Fi9eTG1Dhgxh4sSJVFZWMn/+fGbMmMHChQvJz8/n/vvvZ9CgQXTt2pVZs2Zx3nnn1ew3bNgw/v3vf291POcaWizRNeQPlKWnvm/uUWnTpg0zZszg9ddfZ+rUqYwaNYpbbrmFYcOGbVHukUce4fe//z0rVqxg2rRp9OjRYxtH3D6LFi1i1KhRLFmyhPLy8pp72IcPH87ll1/OmWeeycknn0z37t3Zf//9Offcc6moqOB73/veVomgPi+//DKzZs2qWV69ejVr164F4MQTT6SwsHCrfZYsWULyiLHl5eW88MIL3HXXXRQVFXHggQcyefLkmmsITz31FG+88QYtW7bkgQceYJdddtnieHvuuSdlZWVpx5yOc889l9mzZ1NSUkKvXr045JBDiMfjVFRUcP/99/Puu++yxx57cMkll/Db3/6Wq6++GghaD1988UWDxuJcXRItAvO7hhq3eDzOEUccwfXXX8+9997Ls88+S9++ffn8889Zs2YNAOeccw5lZWW0a9eOqqq6HwyZN28e8XicXXfdlYEDBzJjxoyUdV9yySWMHj2aDz74gAceeKDmPvarrrqKhx56iA0bNjB8+HA++ugjRowYwWuvvUa3bt04++yzeeyxx9I+x+rqaqZPn05ZWRllZWUsXryYNm3aANC6des69yksLNzivvrJkyezcuVKBg0aRHFxMW+88cYWrYhRo0ZRVlbGm2++yUknnbTV8banRdCxY0dWrlxJZWXwNOaiRYvo1q3bVsfMy8vjd7/7HWVlZfz1r39l5cqV9O/fvybh9OnTB0mcdtppTJs2rWa/jRs31pn8nGtom7uGPBE0WnPmzOGTTz6pWS4rK6NXr160atWK8847j9GjR9d8GFZVVdV55wpQ0wc/evRoJDF69GgeffTRmjuQAJ577jm+/PLLLfZbtWpVzQfco48+WrP+008/ZdCgQfz6179m//3356OPPqrpc7/gggs4//zzeeedd9I+z2OOOYZ77rlni/NMpX///ixYsKBmecKECTz00EMsWLCABQsWMH/+fF566SXWr1+fVgyJFkFdr/bt229RVhJHHnkkzzzzDBD8bkaO3HqSvPXr17Nu3ToAXnrpJfLy8hgwYADdunVj1qxZJCZDeumll9h7771r9vv444/ZZ5990orbuZ2xeRjqaI7f7LqGsmHt2rVccsklrFy5kry8PPr27cu4ceOA4LbSa665hn322YeioiIKCws566yz6Nq1KwAbNmxg6NChVFRUkJeXx49+9CMuv/xyALp06cKTTz7JFVdcwbJly4jFYowYMYLjjjtui/rHjBnDqaeeSocOHTjqqKOYP38+EFzMnTp1KrFYjIEDB3L88cfz5JNPcvvtt5Ofn0+bNm22q0Vw9913c/HFFzN48GAqKysZMWIEY8eOrXef1q1b06dPH+bOnUvXrl158cUXt9indevWHHrooUyaNCntOLbHrbfeyumnn87VV1/NvvvuW9PHP3HiREpLS7nhhhtYtmwZxx57LLFYjG7duvH4448D0LVrV6677jpGjBhBfn4+vXr1Yvz48UAwvtXcuXMpKfHpM1z0YuFX9rfmr+CU/bo3+PEVVZ9TVEpKSqy0dMspC2bPnr3FNzXXuPzlL39hxowZW9w51NT95S9/4Z133uHGG2/capv/PbqGVlVt3DBpJkfv3YUR/XdslkZJM8yszm8u3iJwkTvppJNYsWJFtsNoUJWVlfzylylnWXWuQcRj4vqR0XVDeiJwGZF8q2lzcOqpp2Y7BOcaTLO5WNzUurhc8+R/h64pahaJoKCggBUrVvh/QpdVFs5HkHiK3Lmmoll0DXXv3p1FixbV3ObnXLYkZihzrilpFokgPz/fZ4Ryzrkd1Cy6hpxzzu04TwTOOZfjPBE451yOa3JPFktaDny2g7t3Ar5qwHCaAj/n3ODnnBt25px7mVmdjyU3uUSwMySVbusR6+bKzzk3+DnnhqjO2buGnHMux3kicM65HJdriWBctgPIAj/n3ODnnBsiOeecukbgnHNua7nWInDOOVeLJwLnnMtxzTIRSDpO0hxJcyVdVcf2lpKeCre/Kak4C2E2qDTO+XJJsyS9L+lfknplI86GlOqck8p9X5JJavK3GqZzzpJOC/+tZ0p6ItMxNrQ0/rZ7Spoq6d3w7/uEbMTZUCQ9LGmZpA+3sV2S7g5/H+9LGrbTlZpZs3oBceBTYA+gBfAeMKBWmZ8BY8P3pwNPZTvuDJzzkUCr8P1Pc+Gcw3JFwGvAdKAk23Fn4N+5H/Au0CFc3jXbcWfgnMcBPw3fDwAWZDvunTznEcAw4MNtbD8B+Acg4CDgzZ2tszm2CA4A5prZPDMrB54ERtYqMxJ4NHz/DHC0JGUwxoaW8pzNbKqZrQ8XpwNNfazkdP6dAW4EbgU2ZjK4iKRzzhcA95nZNwBmtizDMTa0dM7ZgLbh+3bAFxmMr8GZ2WvA1/UUGQk8ZoHpQHtJu+9Mnc0xEXQDFiYtLwrX1VnGzCqBVUDHjEQXjXTOOdl5BN8omrKU5xw2mXuY2d8zGViE0vl37g/0l/RvSdMlHZex6KKRzjmPAX4oaRHwAnBJZkLLmu39/55Ss5iPwKVP0g+BEuDwbMcSJUkx4C7g7CyHkml5BN1DRxC0+l6TNMjMVmYzqIidAYw3szslHQw8LmkfM6vOdmBNRXNsESwGeiQtdw/X1VlGUh5Bc3JFRqKLRjrnjKT/An4DnGhmmzIUW1RSnXMRsA/wiqQFBH2pE5v4BeN0/p0XARPNrMLM5gMfEySGpiqdcz4PeBrAzP4DFBAMztZcpfX/fXs0x0TwNtBPUm9JLQguBk+sVWYicFb4/hRgioVXYZqolOcsaV/gAYIk0NT7jSHFOZvZKjPrZGbFZlZMcF3kRDMrzU64DSKdv+3nCVoDSOpE0FU0L4MxNrR0zvlz4GgASXsTJILmPG/tRODH4d1DBwGrzGzJzhyw2XUNmVmlpNHAZII7Dh42s5mSbgBKzWwi8EeC5uNcgosyp2cv4p2X5jnfDrQB/hxeF//czE7MWtA7Kc1zblbSPOfJwDGSZgFVwJVm1mRbu2me8y+BByVdRnDh+Oym/MVO0gSCZN4pvO5xHZAPYGZjCa6DnADMBdYD5+x0nU349+Wcc64BNMeuIeecc9vBE4FzzuU4TwTOOZfjPBE451yO80TgnHM5zhOBi5ykKkllkj6UNElS+wY+/oLwnnkkrd1GmUJJr0qKSyqWtCGMaZakseGTyNtTZ4mku8P3R0g6JGnbRZJ+vDPnFB5njKQrUpQZL+mU7Thm8bZGtaxV7mZJC2v/PiWNlnRuuvW5psETgcuEDWY21Mz2IXhu4+IsxHAu8JyZVYXLn5rZUGAwwYiV39ueg5lZqZn9PFw8AjgkadtYM3tsZwPOskkEA77V9jDNfyyfnOOJwGXafwgHyJLUR9KLkmZIel3SXuH6LpL+Ium98HVIuP75sOxMSRduZ71nAn+tvTIcdHAa0Df8tjxFm+ds6BnWe2rYmnlP0mvhuiMk/U3BXBYXAZeFLYzDEt/kJe0l6a1EXeHxPwjf7xe2UGZImqwUo0dKukDS22EMz0pqlbT5vySVSvpY0nfC8nFJt4f7vC/pJ9vzyzKz6XU9rRqOYLtAUl1JwjVRnghcxkiKEwwFkHjqdxxwiZntB1wB/CFcfzfwqpkNIRiXfWa4/tywbAnwc0lpjRgbDk2wh5ktqGNbqzCmD4B7gEfNbDDwpzAOgGuBY8N4tngaOzzmWOB3Yavn9aRtHwEtJPUOV40CnpKUH9Z1Sng+DwM3pziN58xs/zCG2QTj6yQUE3x7/zYwVlJBuH2Vme0P7A9ckBRH4ty7SnohRb11KQUO24H9XCPV7IaYcI1SoaQygpbAbOAlSW0IulMSQ14AtAx/HgX8GCDsylkVrv+5pJPC9z0IBlNLZ/iETsDKWuv6hDEZ8Fcz+4ekx4GTw+2PA7eF7/8NjJf0NPBcGvUle5ogAdwS/hwF7EkwIN5L4bnHgVRjxewj6SagPcFQIZOT6whH2vxE0jxgL+AYYHDS9YN2BL+vjxM7mdkXBEMVbK9lYR2umfBE4DJhg5kNDb99Tya4RjAeWBn206ck6Qjgv4CDzWy9pFcIBhdLq/46yn6abt1mdpGkAwm+cc+QtF+a9QI8RZDsngsOZZ9IGgTMNLODt+M444Hvmdl7ks4mHFguEWLtkAlmr7rEzJITBmqYaVkLCH6nrpnwriGXMWH/8s8JBglbD8yXdCrUzMM6JCz6L4LpNBN93e0IvtF+EyaBvQiGlU633m+AeNhlUp9pbB6A8Ezg9TCGPmb2ppldSzCqZY9a+60hGPa6rro/JRj87RqCpAAwB+isYOx8JOVLGpgitiJgSditdGatbadKiknqQzCl4xyChPvTsDyS+ktqnaKOdPUHUt555JoOTwQuo8zsXeB9gslEzgTOk/QewXWAxBSElwJHhhdWZxDc1fMikCdpNkE3y/TtrPqfwKEpylwCnCPpfeBHYRwAt0v6ILztchrBvLnJJgEnJS4W13Hcp4AfsnnM/HKC4c9vDc+9jKS7jrbhGuBNgm6qj2pt+xx4i2DWuYvMbCPwEDALeCeM+wFq9QDUd41A0m0KRr5sJWmRpDFJm4cDL6WI1zUhPvqoywkKpq28zMx+lO1YmjIF81pc7r/H5sVbBC4nmNk7wNTwziW34zoRtE5cM+ItAuecy3HeInDOuRznicA553KcJwLnnMtxngiccy7HeSJwzrkc9/8BXg2HvwYBf3MAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualization(sgdcModel)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5481367",
   "metadata": {},
   "source": [
    "### Multinomial Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "c464d685",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mislabeled out of a total of 37076 test entries: 3299\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of mislabeled out of a total of %d test entries: %d\" % (urls_test.shape[0], \n",
    "                                                                      (ifSpam_test != nbPredict).sum()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "ddd0971d",
   "metadata": {},
   "outputs": [],
   "source": [
    "successRate = 100.0 * f1_score(ifSpam_test, nbPredict, average='micro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "1bc022f6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Success Rate was calculated as % : 91.10206063221491 with Multinomial Naive Bayes\n"
     ]
    }
   ],
   "source": [
    "print(\"The Success Rate was calculated as % : \" + str(successRate) + \" with Multinomial Naive Bayes\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "05cd7003",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAkyElEQVR4nO3de7wXVb3/8debi9yRmyAKBBhq5jFUQsskxFJEf2Ed89KpzDQyL+k5nUdZWZaezrGsLErrqPFQy2vHS2QooulRT5GiGYpogEiACHLdCIiw9+f3x6y9GWBfvgP7y769n4/HPPbMmjUz66v5aa1ZM/NRRGBmZpl2Td0AM7PmxEHRzCzHQdHMLMdB0cwsx0HRzCzHQdHMLMdB0cyahKTBkh6T9JKkOZIuSeXXSHpZ0mxJ90nqlcqHStok6fm0/DJ3riMlvSBpvqTJkpTK+0iaIWle+tu7wXY1p+cU+/VpH0MHd2zqZlgB8+bu3dRNsAI2VVbwTuUm7c45TjyuW6xaXVlS3Wdnb54eEeNr2ydpIDAwIp6T1AN4FjgVGAT8MSK2Svo+QER8TdJQ4IGIOLSWcz0NfBn4CzANmBwRD0r6AbA6Iq6WdBnQOyK+Vl+bO5T0y/aQoYM78vT0wU3dDCvg5NEnN3UTrIA/vXHHbp9j5epK/jJ9UEl1Ow5c0K+ufRGxDFiW1tdLmgvsHxEP56rNBE6r7xopuPaMiJlp+1ay4PogMBEYm6reAjwO1BsUPXw2s4KCyqgqaQH6SZqVWybVdsbUCzycrKeX93my4FZtmKS/SvpfScemsv2BJbk6S1IZwIAUfAHeAAY09OuaVU/RzJq/AKoo+bbbyogYVV8FSd2Be4BLI6IiV/5NYCtwWypaBgyJiFWSjgTul/TektsdEZIabLiDopkVVkVVo5xHUkeygHhbRNybK/8ccApwfKSJj4jYDGxO689KWgAcCCwluw9ZbVAqA1guaWBELEvD7BUNtcnDZzMrJAi2RFVJS33SDPGvgLkR8eNc+Xjgq8DHImJjrnwfSe3T+nBgBPBqGh5XSDo6nfOzwO/SYVOBs9P62bnyOrmnaGaFBFBZ+vC5PscAnwFekPR8KvsGMBnoBMxIT9bMjIjzgTHAlZK2AFXA+RGxOh13AXAz0IXsHmT1fcirgbslnQssAk5vqFEOimZWWIF7inWKiKeA2h4PmlZH/XvIhtq17ZsF7PSoTkSsAo4v0i4HRTMrJIDKZvR8c2NzUDSzwhpnmqV5clA0s0KCaKx7is2Sg6KZFRIBW1pvTHRQNLOiRGWt8yOtg4OimRUSQJV7imZm27inaGaWZA9vOyiamQFZUNwSrfcNYQdFMyskEJWt+LMJDopmVlhVePhsZgb4nqKZ2Q5Epe8pmpllsi9vOyiamQEQId6J9k3djLJxUDSzwqpa8T3F1tsHNrOyyCZa2pW01EfSYEmPSXpJ0hxJl6TyWhPYKzM5JbyfLemI3LnOTvXnSTo7V36kpBfSMZNTuoJ6OSiaWUHZREspSwO2Al+JiEOAo4ELJR0CXAY8GhEjgEfTNsBJZHlZRgCTgF9AFkSBK4CjgNHAFdWBNNX5Qu648Q01ykHRzAqpnmgpZan3PBHLIuK5tL4emEuWr3kiWeJ60t9T0/pE4NbIzAR6pQx9JwIzImJ1RKwBZgDj076eETEzZQS8NXeuOvmeopkVVln6w9v9JM3Kbd8QETfsWEnSUOBw4C/UncB+f2Bx7rDqpPf1lS+ppbxeDopmVkggtkTJoWNlRIyqr4Kk7mQJqS6NiIr8bb9SE9g3Jg+fzayQxppoAZDUkSwg3hYR96bi5Wnoyw4J7JcCg3OHVye9r698UC3l9XJQNLNCAlEZpS31STPBvwLmRsSPc7vqSmA/FfhsmoU+GliXhtnTgRMk9U4TLCcA09O+CklHp2t9NneuOnn4bGaFNdIbLccAnwFekPR8KvsGdSewnwZMAOYDG4FzACJitaSrgGdSvSsjYnVavwC4GegCPJiWejkomlkhETTKu88R8RTU+RT4Tgns0wzyhXWcawowpZbyWcChRdrloGhmhWQTLX7Nz8yshj8ya2aWBPJHZs3M8txTNDNLsrzPDopmZomcjsDMrFqW4tSzz2ZmQPblbQ+fzcxynLjKzCzJvqfoe4pmZolTnJqZ1cgeyXFP0cwM8LvPZmY7aaRPhzVLDopmVkj26TAPn83MarTme4qttw9sZmWRfSWnXUlLQyRNkbRC0ou5srskPZ+W16q/yi1pqKRNuX2/zB1Ta9J7SX0kzZA0L/3tvVMjduCgaGaFZK/5tStpKcHN7JCgPiLOiIiRETGSLKnVvbndC6r3RcT5ufK6kt5fBjwaESOAR9N2vTx8LtGKpR255pIhrH2zIyiY8OlVfPy8ldx45X7MnNGTjnsFA9+1ma9cu5jue1fyx3t789vr+9ccv3BuZ66b/ncOOHQT82Z34YeXDmHz2+0YPa6CL121FAkq1rTnP88fyvIlezFg0Dt8879fo0evyib81a1Hv/6b+Mp3/kavPu8QwEP3DWbqXcP41Bf+zokTF1Oxdi8Abrn+IGb9qT8jR7/JORe+QoeOVWzd0o5f/exgZs/qB8CHT3id0z83nwixemUnfvjtkVSs26sJf92e1niv+UXEEynn885XyXp7pwPj6m1NLul92q5Oev8gMBEYm6reAjwOfK2+85U1KEoaD/wUaA/cFBFXl/N65dS+QzDp268z4rBNbHyrHReNP5AjxqzniDHr+fw3Xqd9B7jpPwZy58/6c97lyxj3iTWM+8QaIAuI3/38MA44dBMAky8bxKXXLObgIzZy+aeHM+uxHrx/3Hru/nl/Dv/Qes64eAV3/aw/d/08O5ftvspKcdNP38OCV/amS9et/PTWp/jr01mQ+90dw7j3tuHb1a9Yuxff/cooVq/szLuGr+fKyU9z9inH0659FZP+7SW+dMYYKtbtxTkXv8wpp7/G7Tce2BQ/q8kUeKOln6RZue0bIuKGEo89FlgeEfNyZcMk/RWoAC6PiCepP+n9gJTVD+ANYEBDFy3b8FlSe+A64CTgEOAsSYeU63rl1nfAVkYclgW1rt2rGPzuzaxc1pEjx66nffq/lvccuZGVyzrudOxj9/fmwxOzALlqeQc2rm/Pe47ciAQfOW01f3pobwD+PH1vPnJ6loTsI6ev5s+p3HbfmlWdWfBK9s9z08YOLF7Ynb77vF1n/Vf/vjerV3YGYNGr3enUqYoOHbMPZknQqUslEHTttoXVb3beA7+g+aiefS4xxenKiBiVW0oNiABnAXfktpcBQyLicODfgNsl9Sy93RFko/96lfOe4mhgfkS8GhHvAHeSdWVbvDcW78WCF7tw8BEbtyuffkcf3j9u/U71n5jai+NOXQvAqjc60m/glpp9/fbbwso3skC6ZmVH+g7YCkCf/ltZs3LnAGu7r//AjQw/qIJX5vQC4JRPLuLntz3JJZfPpnuPLTvVP2bcGyx4pSdbt7SnsrId133/vVx/+5P8etofGTLsLR6eOninY1q7xppoqYukDsAngLuqyyJic0SsSuvPAguAA6k/6f3yNLyuHmavaOja5QyK+wOLc9v5Lm0NSZMkzZI0681Vzf/+2aYN7bjqvKGcf+VSuvWoqim//acDaN8haobM1V5+riudulQx9OC6eyW1kUBq8P/UrKDOXbbyzauf48YfH8KmDR2Zds+7OO8TY7n40x9izapOnHvJ3O3qDxm+nnMueoWf/VeWJbN9+yom/PM/uPgzx/CZCeNYOK8nn/zcgqb4KU2mOkdLKctu+AjwckTUDIsl7ZNGoEgaTjah8moDSe+nAmen9bNz5XVq8tnniLihumu9T9/m/erQ1i1w1XlDGfeJNXxowrqa8ofv6sPTj/Tkaz9fhHb438Hjv+vF2FO3Bcq++27Zboi98vWO9Ns365307reFVcuzsfiq5R3o1XdrGX9N29O+fRXf+P5zPDZ9P/70+L4ArF3diaoqESEeun8wB753bU39vv03cfkPnuVH3zmMN5Z2A2D4gRUAaVs8+ehA3vNPa3a8VKsWwNZoV9LSEEl3AH8GDpK0RNK5adeZbD90BhgDzE6P6PwPcP4OSe9vAuaT9SCrk95fDXxU0jyyQNvgvEY5J1qWAvlxRb5L2+JEwI+/MoTBIzbzz198s6b8mcd68Nvr+3PNvfPo3HX7nl1VFTzx+1786L75NWV9B2yla49K5j7blYOP2Mgj/9OHiZ/Pznf0CRU8cncfzrh4BY/c3YcPnLgOayzBJd96gcULu3P/7dsmVXr3fZs1q7J7gh8cu5xFC3oA0K37Fr5z7Sxu/vnBzJ3dp6b+qjc7M2TYW/TstZmKtZ04fPRKFr/Wfc/+lGagEWefz6qj/HO1lN1D9ohObfVrTXqfhtvHF2lTOYPiM8AIScPIguGZwKfKeL2ymvN0Nx79nz4Me88mvvSRgwA45+uvc/23BrFls/j6Ge8G4OAjN3DJ97Me/wszu7PPflsY+K53tjvXxf+1hB9eOoR33m7HqOMqau5DnnHRcr53/lAeurMv/ffPHsmxxnHI+9Zw/ISlLJzXg5/95kkge/zmwye8zvADK4gQK5Z1qRkmn3L6IvYbtJGzzpvHWedlk5+XXzya1Ss7c/tN7+YH/z2TrVvbseKNLlx75WFN9ruaxO4PjZs1ZRMyZTq5NAH4CdkjOVMi4nv11R/1vs7x9PS2d9O6JTt59MlN3QQr4E9v3MG6d5bvVkTrfXD/GDfltJLq3nvML56NiFG7c709razPKUbENGBaOa9hZntea+4p+o0WMyvEH5k1M8sJxNaqJn9wpWwcFM2sMCeuMjOrFh4+m5nV8D1FM7MdOCiamSWBqPREi5nZNp5oMTNLwhMtZmbbCwdFM7NqrfuDEA6KZlaYe4pmZkkEVFa13qDYeufVzaxsqlBJS0MkTZG0QtKLubLvSFqaS3o/Ibfv6ynh/SuSTsyVj09l8yVdlisfJukvqfwuSQ3monVQNLNCgmz4XMpSgpvZlrg+79pc0vtpACkb6JnAe9Mx10tq30Dm0O+nc70bWAOcu+OFduSgaGYFNV7iqoh4AljdYMXMRODOlNVvIVk+ltHUkTk0JbEaR5bPBeAW4NSGLuKgaGaFRZS2AP2qs3WmZVKJl7hI0uw0vO6dyurKEFpXeV9gbURs3aG8Xp5oMbPCCsw+r9yFdAS/AK4iG6lfBfwI+HzBc+wyB0UzKySbfS7fIDMillevS7oReCBt1pchtLbyVUAvSR1Sb7GkjKIePptZYQWGz4VJGpjb/DhQPTM9FThTUqeUJXQE8DS5zKFpdvlMYGpkWfkeA6qzbJ0N/K6h67unaGaFNdbD25LuAMaS3XtcAlwBjJU0kmz4/BrwxeyaMUfS3cBLwFbgwoioTOe5CJjOtsyhc9IlvgbcKek/gL8Cv2qoTQ6KZlZIUPLjNg2fK+KsWorrDFwpTfJOqZLryhwaEa+SzU6XzEHRzAorX7b4puegaGbFBEQrfs3PQdHMCvMHIczMcnZ1ZrklqDMoSvoZ9dw6iIgvl6VFZtasVb/73FrV11OctcdaYWYtRwBtMShGxC35bUldI2Jj+ZtkZs1dax4+N/hGi6QPSHoJeDltv0/S9WVvmZk1UyKqSltaolJe8/sJcCLZe4RExN+AMWVsk5k1d1Hi0gKVNPscEYuzT5PVqCxPc8ys2Yu2O9FSbbGkDwIhqSNwCTC3vM0ys2athfYCS1HK8Pl84EKyjzO+DoxM22bWZqnEpeVpsKcYESuBf9kDbTGzlqKqqRtQPqXMPg+X9HtJb6asW7+TNHxPNM7MmqHq5xRLWVqgUobPtwN3AwOB/YDfAneUs1Fm1ryV8yOzTa2UoNg1In4dEVvT8hugc7kbZmbNWCM9klNH3udrJL2cElfdJ6lXKh8qaVMuH/Qvc8ccKemFlN95csrkh6Q+kmZImpf+9t6pETuoMyimk/UBHpR0WWrQuyR9lVo+5mhmbUjjDZ9vZue8zzOAQyPiMODvwNdz+xbk8kGfnyv/BfAFshQFI3LnvAx4NCJGAI+m7XrVN9HyLFmsr/5lX8ztix0aamZtiBppaBwRT0gaukPZw7nNmWzLsVJ7W7KcLj0jYmbavpUsv/ODZLmix6aqtwCPk6UoqFN97z4Pq+9AM2ujQlD6K3z9JOU/LnNDRNxQ4GqfB+7KbQ+T9FegArg8Ip4ke1xwSa5OPr/zgIhYltbfAAY0dMGS3miRdChwCLl7iRFxaynHmlkrVHpPcVfyPgMg6ZtkCapuS0XLgCERsUrSkcD9kt5b6vkiIqSG+7gNBkVJV5B1Pw8hu5d4EvAU4KBo1laVeWZZ0ueAU4DjU6pSImIzsDmtPytpAXAgWS7nQbnD8/mdl0saGBHL0jB7RUPXLmX2+TTgeOCNiDgHeB+wdyk/zMxaqTJ+EELSeOCrwMfynyuUtI+k9ml9ONmEyqtpeFwh6eg06/xZtuV3nkqW7xkaMe/zpoiokrRVUk+ySDu4tJ9nZq1OI35kto68z18HOgEz0pM1M9NM8xjgSklbyN6pOT8iVqdTXUA2k92FbILlwVR+NXC3pHOBRcDpDbWplKA4Kz0ndCPZjPRbwJ9LOM7MWqlGnH0uOe9zRNwD3FPHvlnAobWUryIb6ZaslHefL0irv5T0ENnU9+wiFzGzVqaFvq1SivoSVx1R376IeK48TTKz5q6xeorNUX09xR/Vsy+AcY3cFv4+uysn7jeysU9rZbTw6iFN3QQrYPPkvRrnRC30Yw+lqO/h7eP2ZEPMrIVowakGSlHSw9tmZttxUDQz20at+COzDopmVlwr7imW8uVtSfq0pG+n7SGSRpe/aWbWHClKX1qiUl7zux74AFD9kOV64LqytcjMmr9WnI6glOHzURFxRPpcDxGxRlIjzeubWYvUQnuBpSglKG5JL2EHZC9l06pzeZlZQ1rq0LgUpQTFycB9QH9J3yP7as7lZW2VmTVf0cZnnyPiNknPkr1ULeDUiJhb9paZWfPVlnuKkoYAG4Hf58si4h/lbJiZNWNtOSgCf2BbAqvOwDDgFaDkz4CbWevSpu8pRsQ/5bfT13MuqKO6mVmLVspzittJnww7qgxtMbOWopHSEUiaImmFpBdzZbUmsE8vkkxOCe9n5z9vKOnsVH+epLNz5UdKeiEdMzmlK6hXKW+0/Ftu+XdJtwOvN/xzzaxVSrPPpSwluJltieur1ZXA/iS2JbufBPwCsiBKlsbgKGA0cEV1IE11vpA7bsdr7aSUnmKP3NKJ7B7jxBKOM7PWqpF6ihHxBLB6h+KJZInrSX9PzZXfGpmZQK+Uoe9EYEZErI6INcAMYHza1zMiZqaMgLfmzlWneu8ppoe2e0TEvzf888ysLRCFJlr6SZqV274hIm5o4Ji6EtjvDyzO1atOel9f+ZJayutVXzqCDhGxVdIxDZ3EzNqY0oPiyogYtcuXKTGBfWOqr6f4NHAE8LykqcBvgQ3VOyPi3jK3zcyao/J/AaeuBPZL2T69cnXS+6VkaVLz5Y+n8kG11K9XKfcUOwOryHKynAL8v/TXzNqqqhKXXVNXAvupwGfTLPTRwLo0zJ4OnCCpd5pgOQGYnvZVSDo6zTp/NneuOtXXU+wv6d+AF9n28Ha1Vvzoppk1pLF6ipLuIOvl9ZO0hGwWua4E9tOACcB8srfszgGIiNWSrgKeSfWujIjqyZsLyGa4uwAPpqVe9QXF9kB3tg+G1RwUzdqyRooAEXFWHbt2SmCfZpAvrOM8U4AptZTPAg4t0qb6guKyiLiyyMnMrA1ow9n8WuZnc82s7Nrqu887dV/NzIC22VPM3ag0M9tOm/7IrJnZdtrwPUUzs52I1j3h4KBoZsW5p2hmtk1bnX02M6udg6KZWdLWU5yame3EPUUzs218T9HMLM9B0cxsG/cUzcyqBbvzAdlmz0HRzAopmLiqxSklHYGZ2fYaIcWppIMkPZ9bKiRdKuk7kpbmyifkjvl6Smz/iqQTc+XjU9l8SZfVfsXSuKdoZoUpdr+rGBGvACOhJp3yUuA+sjQD10bED7e7pnQIcCbwXmA/4BFJB6bd1wEfJUtj+oykqRHx0q60y0HRzIopz1dyjgcWRMSiLMdUrSYCd0bEZmChpPnA6LRvfkS8CiDpzlR3l4Kih89mVpiitIUsIdWs3DKpjlOeCdyR275I0mxJU1KGPqg/6X1t5bvEQdHMClNVaQuwMiJG5ZYbdjqXtBfwMbLc8gC/AA4gG1ovA360R35U4uGzmRXXuMPnk4DnImI5QPVfAEk3Ag+kzaXA4Nxx+eT2dZUX5p6imRVT4tC5wGM7Z5EbOksamNv3cbLc8wBTgTMldZI0DBgBPE2W73mEpGGp13lmqrtL3FM0s+IaqacoqRvZrPEXc8U/kDQyXeW16n0RMUfS3WQTKFuBCyOiMp3nImA6Wb76KRExZ1fb5KBoZoU05sPbEbEB6LtD2Wfqqf894Hu1lE8DpjVGmxwUzawwVbXeV1ocFM2sGGfzs/oMOuBtvvHLRTXb+w55h19fsy/vGbWBQQdsBqBbz0o2VLTngo8eRI/eW/nWDa9x4MhNzLi7N9d9c1BTNb3V+88PPsZx+y9i1dtdOOX3ZwAw/l0LuPh9szhg7zWcNu0TvLiqf039Lx76HKe9+2UqQ/zHMx/iqdcHM6znWn4yZkZNncHdK/jp397PLXMP4ydjZjCs51oAeuy1mfXvdGLiA5/co7+xqfjL27tA0hTgFGBFRBxarus0tSULOnPBRw8CoF274LbnXuL/Htyb+27ap6bOpG+/zob12UT/O2+LW67Zl6EHvc3Qg99ukja3FffOP4jfvHwoPzjmjzVl89b24aLHT+TKo/93u7oH7L2ak4cuYMLUMxjQdQM3f/QBTrj/TBZW9KoJdO1UxZOn/ZoZ/xgGwKVPfLTm+MuO/BPrt+y1B35VM9GKe4rlfCTnZmB8Gc/f7Iw89i2WLdqLFUvz/3EEYz62lsfuzx7K37ypPXOe7s47m/00VLnNWrEf6zZ32q5swbreLKzotVPdjwx+jT+8dgBbqtqz5K2eLFrfk8P6rtiuzgf2Xco/1vfk9Q09djg6OGnoAh5Y+O5G/gXNVyM/ktOslO2/zIh4AlhdrvM3R2MnruHx+3tvV3boURtY82YHXl/YqY6jrDkY0HUDyzZ0r9l+Y0N3BnTdsF2dk4fN5w8LR+x07Kj+y1i5qSuL1vcqdzObhwAiSltaoCbvrkiaVP1e5BY2N3VzdlmHjlUcfUIFT/x+7+3Kjzt1LY/f36tpGmWNpmO7So4ftIgHFw3fad8pw+bzh9faTi8RCr3m1+I0eVCMiBuq34vsSMvtTb1/3Hrmv9CFtSs71pS1ax8cM2Ed/zu1V9M1zEqyfGM3BnZ7q2Z7325vsXxjt5rtMfv/gzmr+7Hq7a7bHddeVZwwZCF/eO2APdbWplb9nKKHz1avsaeu3WnofMSx61k8vxMrl7WhG/At1KOLh3Ly0AV0bFfJoO4VDO2xjtm5melThs6v9Z7hBwcu4dV1vVi+sftO+1qtUofOLXT47EdyGkGnLpUccex6fvrV7R+v+fDE2ofOt/zlJbp1r6LDXsEHTqzgG2cN5x/zOu+h1rYdPz72EUYPeJ3end/miX/+NZP/Nop1mzvzrdFP0afzJm4Y9yBz1/Tl3EdOYf66PkxbNJwHJ97F1irx3b8cS1VkfYYuHbbwwf2W8K2ZY3a6xslD5/NAGxs6Q8vtBZZCUaZoLukOYCzQD1gOXBERv6rvmJ7qE0fp+LK0x8pj4dUfaOomWAFLJl/L20sW1/kV11L06DUoDh9zSUl1n/z9V5+NiFG7c709rWw9xYg4q1znNrOm1Zp7ih4+m1kxAVS23qjooGhmhbmnaGaW10JnlkvhoGhmhbXmnqKfUzSzYupKfF/b0gBJr0l6ISW9n5XK+kiaIWle+ts7lUvS5JTwfrakI3LnOTvVnyfp7N35eQ6KZlaIAFVGSUuJjouIkblHdy4DHo2IEcCjaRuyBFcj0jKJLOsfkvoAVwBHkeWBviKXFrUwB0UzK0wRJS27aCJwS1q/BTg1V35rZGYCvVKSqxOBGRGxOiLWADPYjS90OSiaWTHFhs/9qj/4kpZJtZztYUnP5vYNiIhlaf0NYEBaryvpfV3lu8QTLWZWUKH3mlc28EbLhyJiqaT+wAxJL293pYiQ9uy0jnuKZlZYY30lJyKWpr8rgPvI7gkur879nP5Wf+13KbUnva+rfJc4KJpZcY3wlRxJ3ST1qF4HTiBLfD8VqJ5BPhv4XVqfCnw2zUIfDaxLw+zpwAmSeqcJlhNS2S7x8NnMigmKzCzXZwBwnyTIYtHtEfGQpGeAuyWdCywCTk/1pwETgPnARuAcgIhYLekq4JlU78qI2OWv/jsomllxjRATI+JV4H21lK8CdvpcVmSf9LqwjnNNAabsfqscFM1sF+zG4zbNnoOimRXnoGhmlgTQQpNSlcJB0cwKEbv1tkqz56BoZsVVtd6uooOimRXj4bOZ2fY8fDYzy3NQNDOr1nIT3ZfCQdHMinE2PzOz7fmeoplZnoOimVkSQJWDoplZ4okWM7PtOSiamSUBVLbeV1qcjsDMCgqIqtKWekgaLOkxSS9JmiPpklT+HUlLJT2flgm5Y74uab6kVySdmCsfn8rmS7qstuuVyj1FMyuucYbPW4GvRMRzKVfLs5JmpH3XRsQP85UlHQKcCbwX2A94RNKBafd1wEfJ0ps+I2lqRLy0K41yUDSzYhpp9jklnVqW1tdLmkv9+ZonAndGxGZgoaT5ZNn/AOan9AZIujPV3aWg6OGzmRVXeja/fpJm5ZZJtZ1O0lDgcOAvqegiSbMlTUkZ+qDupPd1le8SB0UzK670oLgyIkbllht2PJWk7sA9wKURUQH8AjgAGEnWk/zRnvthHj6bWVERUFnZKKeS1JEsIN4WEfdmp4/luf03Ag+kzfqS3tdVXph7imZWXOk9xTopS/j8K2BuRPw4Vz4wV+3jwItpfSpwpqROkoYBI4CnyfI9j5A0TNJeZJMxU3f1p7mnaGbFNc7s8zHAZ4AXJD2fyr4BnCVpJNmUzmvAF7NLxhxJd5NNoGwFLoyISgBJFwHTgfbAlIiYs6uNclA0s4KisWafnwJUy65p9RzzPeB7tZRPq++4IhwUzayYgGjgweyWzEHRzIprxa/5OSiaWTERTnFqZrYdfyXHzGybcE/RzKyaPzJrZraN0xGYmW0TQDTSa37NkYOimRUT0eAHZFsyB0UzKyw8fDYzy2nFPUVFM5pFkvQmsKip21EG/YCVTd0IK6S1/jt7V0TsszsnkPQQ2T+fUqyMiPG7c709rVkFxdZK0qyIGNXU7bDS+d9Z2+XvKZqZ5TgompnlOCjuGTvlpbBmz//O2ijfUzQzy3FP0cwsx0HRzCzHQbGMJI2X9Iqk+ZIua+r2WMNS8vUVkl5suLa1Rg6KZSKpPXAdcBJwCFmGskOatlVWgpuBFvWwsTUuB8XyGQ3Mj4hXI+Id4E5gYhO3yRoQEU8Aq5u6HdZ0HBTLZ39gcW57SSozs2bMQdHMLMdBsXyWAoNz24NSmZk1Yw6K5fMMMELSMEl7AWcCU5u4TWbWAAfFMomIrcBFwHRgLnB3RMxp2lZZQyTdAfwZOEjSEknnNnWbbM/ya35mZjnuKZqZ5TgompnlOCiameU4KJqZ5TgompnlOCi2IJIqJT0v6UVJv5XUdTfOdbOk09L6TfV9rELSWEkf3IVrvCZpp6xvdZXvUOetgtf6jqR/L9pGsx05KLYsmyJiZEQcCrwDnJ/fKWmX8nhHxHkR8VI9VcYChYOiWUvkoNhyPQm8O/XinpQ0FXhJUntJ10h6RtJsSV8EUObn6fuOjwD9q08k6XFJo9L6eEnPSfqbpEclDSULvv+aeqnHStpH0j3pGs9IOiYd21fSw5LmSLoJUEM/QtL9kp5Nx0zaYd+1qfxRSfuksgMkPZSOeVLSwY3yT9Ms2aWehTWt1CM8CXgoFR0BHBoRC1NgWRcR75fUCfg/SQ8DhwMHkX3bcQDwEjBlh/PuA9wIjEnn6hMRqyX9EngrIn6Y6t0OXBsRT0kaQvbWznuAK4CnIuJKSScDpbwN8vl0jS7AM5LuiYhVQDdgVkT8q6Rvp3NfRJZQ6vyImCfpKOB6YNwu/GM0q5WDYsvSRdLzaf1J4Fdkw9qnI2JhKj8BOKz6fiGwNzACGAPcERGVwOuS/ljL+Y8Gnqg+V0TU9V3BjwCHSDUdwZ6SuqdrfCId+wdJa0r4TV+W9PG0Pji1dRVQBdyVyn8D3Juu8UHgt7lrdyrhGmYlc1BsWTZFxMh8QQoOG/JFwMURMX2HehMasR3tgKMj4u1a2lIySWPJAuwHImKjpMeBznVUj3TdtTv+MzBrTL6n2PpMB74kqSOApAMldQOeAM5I9xwHAsfVcuxMYIykYenYPql8PdAjV+9h4OLqDUkj0+oTwKdS2UlA7wbaujewJgXEg8l6qtXaAdW93U+RDcsrgIWSPpmuIUnva+AaZoU4KLY+N5HdL3wuJV/6b7IRwX3AvLTvVrIvwWwnIt4EJpENVf/GtuHr74GPV0+0AF8GRqWJnJfYNgv+XbKgOodsGP2PBtr6ENBB0lzgarKgXG0DMDr9hnHAlan8X4BzU/vm4BQP1sj8lRwzsxz3FM3MchwUzcxyHBTNzHIcFM3MchwUzcxyHBTNzHIcFM3Mcv4/4DVjPo0ECKMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualization(nbModel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f4d8c38c",
   "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": "3.9.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}