fake_news_classification_machine_learning_approach.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fake New Classification using news headline\n",
    "\n",
    "In this notebook I have demonstrated the machine learning approach to detect fake news from news headline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import itertools\n",
    "import string\n",
    "from nltk.stem import WordNetLemmatizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer, HashingVectorizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import PassiveAggressiveClassifier,LogisticRegression\n",
    "from sklearn.naive_bayes import MultinomialNB,GaussianNB\n",
    "from sklearn import metrics\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    See full source and example: \n",
    "    http://scikit-learn.org/stable/auto_examples/model_selection/plot_confusion_matrix.html\n",
    "    \n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, cm[i, j],\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('fake_news.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>Unnamed: 0</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>As U.S. budget fight looms, Republicans flip t...</td>\n",
       "      <td>WASHINGTON (Reuters) - The head of a conservat...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>U.S. military to accept transgender recruits o...</td>\n",
       "      <td>WASHINGTON (Reuters) - Transgender people will...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>Senior U.S. Republican senator: 'Let Mr. Muell...</td>\n",
       "      <td>WASHINGTON (Reuters) - The special counsel inv...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>FBI Russia probe helped by Australian diplomat...</td>\n",
       "      <td>WASHINGTON (Reuters) - Trump campaign adviser ...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>Trump wants Postal Service to charge 'much mor...</td>\n",
       "      <td>SEATTLE/WASHINGTON (Reuters) - President Donal...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                              title  \\\n",
       "0           0  As U.S. budget fight looms, Republicans flip t...   \n",
       "1           1  U.S. military to accept transgender recruits o...   \n",
       "2           2  Senior U.S. Republican senator: 'Let Mr. Muell...   \n",
       "3           3  FBI Russia probe helped by Australian diplomat...   \n",
       "4           4  Trump wants Postal Service to charge 'much mor...   \n",
       "\n",
       "                                                text label  \n",
       "0  WASHINGTON (Reuters) - The head of a conservat...  TRUE  \n",
       "1  WASHINGTON (Reuters) - Transgender people will...  TRUE  \n",
       "2  WASHINGTON (Reuters) - The special counsel inv...  TRUE  \n",
       "3  WASHINGTON (Reuters) - Trump campaign adviser ...  TRUE  \n",
       "4  SEATTLE/WASHINGTON (Reuters) - President Donal...  TRUE  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(['Unnamed: 0'],axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>as u.s. budget fight looms, republicans flip t...</td>\n",
       "      <td>WASHINGTON (Reuters) - The head of a conservat...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>u.s. military to accept transgender recruits o...</td>\n",
       "      <td>WASHINGTON (Reuters) - Transgender people will...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>senior u.s. republican senator: 'let mr. muell...</td>\n",
       "      <td>WASHINGTON (Reuters) - The special counsel inv...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>fbi russia probe helped by australian diplomat...</td>\n",
       "      <td>WASHINGTON (Reuters) - Trump campaign adviser ...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>trump wants postal service to charge 'much mor...</td>\n",
       "      <td>SEATTLE/WASHINGTON (Reuters) - President Donal...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  as u.s. budget fight looms, republicans flip t...   \n",
       "1  u.s. military to accept transgender recruits o...   \n",
       "2  senior u.s. republican senator: 'let mr. muell...   \n",
       "3  fbi russia probe helped by australian diplomat...   \n",
       "4  trump wants postal service to charge 'much mor...   \n",
       "\n",
       "                                                text label  \n",
       "0  WASHINGTON (Reuters) - The head of a conservat...  TRUE  \n",
       "1  WASHINGTON (Reuters) - Transgender people will...  TRUE  \n",
       "2  WASHINGTON (Reuters) - The special counsel inv...  TRUE  \n",
       "3  WASHINGTON (Reuters) - Trump campaign adviser ...  TRUE  \n",
       "4  SEATTLE/WASHINGTON (Reuters) - President Donal...  TRUE  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# converting to lowercase\n",
    "df['title']=df['title'].str.lower()\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Removing Stopwords"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.corpus import stopwords\n",
    "stop = set(stopwords.words('english'))\n",
    "punctuation = list(string.punctuation)\n",
    "stop.update(punctuation)\n",
    "\n",
    "# Removing stop words which are unneccesary from headline news\n",
    "def remove_stopwords(text):\n",
    "    final_text = []\n",
    "    for i in text.split():\n",
    "        if i.strip().lower() not in stop:\n",
    "            final_text.append(i.strip())\n",
    "    return \" \".join(final_text)\n",
    "\n",
    "df['title']=df['title'].apply(remove_stopwords)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>us budget fight looms republicans flip fiscal ...</td>\n",
       "      <td>WASHINGTON (Reuters) - The head of a conservat...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>us military accept transgender recruits monday...</td>\n",
       "      <td>WASHINGTON (Reuters) - Transgender people will...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>senior us republican senator let mr mueller job</td>\n",
       "      <td>WASHINGTON (Reuters) - The special counsel inv...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>fbi russia probe helped australian diplomat ti...</td>\n",
       "      <td>WASHINGTON (Reuters) - Trump campaign adviser ...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>trump wants postal service charge much more am...</td>\n",
       "      <td>SEATTLE/WASHINGTON (Reuters) - President Donal...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  us budget fight looms republicans flip fiscal ...   \n",
       "1  us military accept transgender recruits monday...   \n",
       "2    senior us republican senator let mr mueller job   \n",
       "3  fbi russia probe helped australian diplomat ti...   \n",
       "4  trump wants postal service charge much more am...   \n",
       "\n",
       "                                                text label  \n",
       "0  WASHINGTON (Reuters) - The head of a conservat...  TRUE  \n",
       "1  WASHINGTON (Reuters) - Transgender people will...  TRUE  \n",
       "2  WASHINGTON (Reuters) - The special counsel inv...  TRUE  \n",
       "3  WASHINGTON (Reuters) - Trump campaign adviser ...  TRUE  \n",
       "4  SEATTLE/WASHINGTON (Reuters) - President Donal...  TRUE  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spec_chars = [\"-\",\"!\",'\"',\"#\",\"%\",\"&\",\"'\",\"(\",\")\",\n",
    "              \"*\",\"+\",\",\",\"-\",\".\",\"/\",\":\",\";\",\"<\",\n",
    "              \"=\",\">\",\"?\",\"@\",\"[\",\"\\\\\",\"]\",\"^\",\"_\",\n",
    "              \"`\",\"{\",\"|\",\"}\",\"~\",\"–\"]\n",
    "for char in spec_chars:\n",
    "    df['title'] = df['title'].str.replace(char, '')\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### shuffle dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df =df.reindex(np.random.permutation(df.index))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>title</th>\n",
       "      <th>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>36745</th>\n",
       "      <td>nc event change name word ‘dixie’ racist ‘offe...</td>\n",
       "      <td>This is political correctness run amuck with t...</td>\n",
       "      <td>FAKE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43715</th>\n",
       "      <td>us middle class still suffering rockefellerkis...</td>\n",
       "      <td>Dick Eastman The Truth HoundWhen Henry Kissin...</td>\n",
       "      <td>FAKE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15378</th>\n",
       "      <td>zimbabwes axed vp safe traveling south africa ...</td>\n",
       "      <td>JOHANNESBURG (Reuters) - Zimbabwe s former vic...</td>\n",
       "      <td>TRUE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38927</th>\n",
       "      <td>in disgraced democrat harry reid funneled mass...</td>\n",
       "      <td>The Defense Department secretly set up a progr...</td>\n",
       "      <td>FAKE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21792</th>\n",
       "      <td>trump fans crying milk hero breaks yet another...</td>\n",
       "      <td>Trump made a bad decision, or so his ardent fo...</td>\n",
       "      <td>FAKE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   title  \\\n",
       "36745  nc event change name word ‘dixie’ racist ‘offe...   \n",
       "43715  us middle class still suffering rockefellerkis...   \n",
       "15378  zimbabwes axed vp safe traveling south africa ...   \n",
       "38927  in disgraced democrat harry reid funneled mass...   \n",
       "21792  trump fans crying milk hero breaks yet another...   \n",
       "\n",
       "                                                    text label  \n",
       "36745  This is political correctness run amuck with t...  FAKE  \n",
       "43715   Dick Eastman The Truth HoundWhen Henry Kissin...  FAKE  \n",
       "15378  JOHANNESBURG (Reuters) - Zimbabwe s former vic...  TRUE  \n",
       "38927  The Defense Department secretly set up a progr...  FAKE  \n",
       "21792  Trump made a bad decision, or so his ardent fo...  FAKE  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df.label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df['title']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y,test_size=0.25, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Baseline model using count vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "count_vectorizer = CountVectorizer(stop_words='english')\n",
    "count_train = count_vectorizer.fit_transform(X_train)\n",
    "count_test = count_vectorizer.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.940\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mnb = MultinomialNB()\n",
    "mnb.fit(count_train, y_train)\n",
    "pred = mnb.predict(count_test)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.943\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVcAAAEmCAYAAADWT9N8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3deZxd8/3H8dd7EonEIImIJYtYUmsJInaCNIJoUFtRsab6o5vSKtpQtMqv1qDV2ooS/bWIpYitSm0JsYSQRJFUJJNVNjGZfH5/nDNxE5mZO8m9c8/ceT89zmPu+Z7vOedzMvLJ937P93yPIgIzMyusilIHYGZWjpxczcyKwMnVzKwInFzNzIrAydXMrAicXM3MisDJ1eolqZ2khyTNlfTX1TjO8ZKeKGRspSJpb0nvlToOyzZ5nGt5kHQccDawFTAPGAtcFhHPr+ZxvwN8H9gjIpasdqAZJymAXhExsdSxWPPmlmsZkHQ2cA3wa2ADoAdwIzC4AIffBHi/JSTWfEhqXeoYrJmICC/NeAHWBeYDR9VTpy1J8v0kXa4B2qbb+gFTgJ8A04GpwMnptouBL4Dq9BynAhcBd+UcuycQQOt0/STgA5LW83+A43PKn8/Zbw/gVWBu+nOPnG3PApcAL6THeQLoXMe11cb/05z4DwMOBt4HZgHn59TvC7wIzEnrDgfapNueS69lQXq9x+Qc/2fAp8CdtWXpPpun59gpXd8YmAH0K/X/G15Ku7jl2vztDqwJ3F9PnQuA3YDewA4kCebCnO0bkiTpriQJ9AZJHSNiGElreEREVEbELfUFImkt4DrgoIhYmySBjl1JvU7AI2nd9YCrgEckrZdT7TjgZKAL0AY4p55Tb0jyZ9AV+CXwR+AEYGdgb+CXkjZL69YAPwY6k/zZHQD8D0BE7JPW2SG93hE5x+9E0oofmnviiJhEknjvltQeuA24PSKerSdeawGcXJu/9YAZUf/X9uOBX0XE9IioImmRfidne3W6vToiHiVptW25ivEsBbaT1C4ipkbEuJXUOQSYEBF3RsSSiLgHGA8cmlPntoh4PyIWAfeR/MNQl2qS/uVq4F6SxHltRMxLzz8O2B4gIsZExEvpeT8E/gDsm8c1DYuIxWk8y4mIPwITgJeBjUj+MbMWzsm1+ZsJdG6gL3Bj4KOc9Y/SsmXHWCE5LwQqGxtIRCwg+Sp9BjBV0iOStsojntqYuuasf9qIeGZGRE36uTb5TcvZvqh2f0lfk/SwpE8lfUbSMu9cz7EBqiLi8wbq/BHYDrg+IhY3UNdaACfX5u9F4HOSfsa6fELylbZWj7RsVSwA2uesb5i7MSIej4hvkLTgxpMknYbiqY3pv6sYU2PcRBJXr4hYBzgfUAP71DukRlIlST/2LcBFabeHtXBOrs1cRMwl6We8QdJhktpLWkPSQZKuSKvdA1woaX1JndP6d63iKccC+0jqIWld4Oe1GyRtIOmbad/rYpLuhZqVHONR4GuSjpPUWtIxwDbAw6sYU2OsDXwGzE9b1d9bYfs0YLOv7FW/a4ExEXEaSV/y71c7Smv2nFzLQERcRTLG9UKgCpgMnAU8kFa5FBgNvAm8BbyWlq3KuUYBI9JjjWH5hFhBMurgE5I76PuS3ixa4RgzgUFp3Zkkd/oHRcSMVYmpkc4huVk2j6RVPWKF7RcBd0iaI+nohg4maTAwkKQrBJLfw06Sji9YxNYs+SECM7MicMvVzKwInFzNzIrAydXMrAicXM3MiqCsJ6FQ63ahNmuXOgxrhB237lHqEKwRPvroQ2bMmNHQOOFGabXOJhFLvvIg3ErFoqrHI2JgIc9fKOWdXNusTdstGxxNYxnywsvDSx2CNcKeu/Yp+DFjyaK8/95+PvaGhp6uK5myTq5m1hwJ1Px7LJ1czSxbBFS0KnUUq83J1cyyRwXtxi0JJ1czyxh3C5iZFYdbrmZmBSbccjUzKzy55WpmVhQeLWBmVmi+oWVmVnjC3QJmZkXhlquZWaG5W8DMrPAEtPINLTOzwnOfq5lZoblbwMysONxyNTMrgjJouTb/KzCz8iLlv+R1OH0o6S1JYyWNTss6SRolaUL6s2NaLknXSZoo6U1JO+UcZ0haf4KkIQ2d18nVzLKnolV+S/72i4jeEVH7XprzgKciohfwVLoOcBDQK12GAjdBkoyBYcCuQF9gWG1CrvMSGhOdmVnxpTe08llW3WDgjvTzHcBhOeV/jsRLQAdJGwEHAqMiYlZEzAZGAfW+GNHJ1cyyJ/9ugc6SRucsQ1dytACekDQmZ/sGETEVIP3ZJS3vCkzO2XdKWlZXeZ18Q8vMsqVx87nOyPmqX5c9I+ITSV2AUZLGN3D2FUU95XVyy9XMMqaw3QIR8Un6czpwP0mf6bT06z7pz+lp9SlA95zduwGf1FNeJydXM8ueAo0WkLSWpLVrPwMDgLeBkUDtHf8hwIPp55HAiemogd2AuWm3wePAAEkd0xtZA9KyOrlbwMyyp3CTZW8A3K8kEbcG/hIRj0l6FbhP0qnAx8BRaf1HgYOBicBC4GSAiJgl6RLg1bTeryJiVn0ndnI1s2xR4R5/jYgPgB1WUj4TOGAl5QGcWcexbgVuzffcTq5mlj1+/NXMrPDk5GpmVljJW16cXM3MCktCFU6uZmYF55armVkROLmamRWBk6uZWaGJlT/J38w4uZpZpgi55WpmVgwVFc1/2hMnVzPLHLdczcwKzX2uZmbF4ZarmVmB+YaWmVmROLmamRWa8NwCZmbF4JarmVkROLmamRWYb2iZmRVL88+tTq5ZMf6Ri5m3YDE1S5eypGYpex1/BRd892BOOWIPqmbPB2DY8JE8/vw77L/rVlzyg2/SZo3WfFG9hPOveYB/vvo+AGu0bsXV5x3NPn16sXTpUi664WEeeGpsKS+t7H3++ef0328fvli8mCU1Szj8iCP5xbCLuemG4Qy//ho+mDSJyVOr6Ny5MwBX/e5KRvzlbgCW1Cxh/LvvMnlqFZ06dSrlZWSH/PirFdjAodcyc86C5cquv+sZrrnzqeXKZs6Zz5E/+gNTq+ayzeYb8dCNZ7L5gRcC8LPTDqRq1jy2P+xXSKLTuu2bLP6Wqm3btjw26mkqKyuprq5m/333YsCBB7H7Hnty8CGDGNC/33L1z/7JuZz9k3MBeOThh7j+2qudWFfgbgEriTfem7Ls8zuTptK2zRrLWrFDBu/ODodfAkBEfCVZW+FJorKyEoDq6mqWVFcjid477tjgvveNuIejj/l2sUNsfpp/bqX5t73LRETw0I1n8cLdP+WUI/ZcVn7Gsfvwyoif8/thx9Nh7XZf2e/w/r15473JfFG9hHUrk+3DzhzEv//yM+6+4hS6dFq7ya6hJaupqWHXnXvTY+Mu7N//G/TdddcG91m4cCGjHn+Mw474VhNE2LxIymvJsqImV0k1ksbmLD1ztl0r6b+SKnLKTpI0PP1cIekOSbcq8aGkt3KOdV0xY29q+598NXsc91sOO+tGvnvM3uy50+b88a//YptDL2LXYy/n0xmfcfnZRyy3z9abbcilPxjMWZfeC0Dr1hV027AjL479gD2O+y0vv/khv/nx4aW4nBanVatWvDxmLBM/nMLoV19h3NtvN7jPIw8/xO577OkugRXkm1hbdHIFFkVE75zlQ0gSJ3A4MBnYZ8WdlPyp/R5YAzgtIiLdtF/OsX5Q5Nib1NSquQBUzZ7PyKffZJdtezJ91jyWLg0iglv//gJ9tttkWf2uXTow4qqhnPaLO/nPlBkAzJyzgAWLFvPg028A8PdRr9F76+5NfzEtWIcOHdhn33488cRjDdb96333cpS7BFbKyXXV7Qe8DdwErOz/rmuB9YATI2JpUwZWCu3XbENl+7bLPvfffSvGTfqEDTuvs6zO4P134J1JUwFYt7Idf7/+DH55/UhefOOD5Y716HNvs0+fXgD067sl4z+Y2kRX0XJVVVUxZ84cABYtWsTTTz3JlltuVe8+c+fO5fnn/smh3xzcFCE2O6pQXkuWFfuGVjtJteOA/hMRtd9Rvw3cAzwI/FrSGhFRnW47DngX6BcRS1Y43jOSatLPd0TE1SueUNJQYCgAa1QW7kqKqMt6azPiqtMBaN2qFSP+MZpR/36XWy45ke237EZE8NHUWXz/0nuApB928+7rc97pAznv9IEAHPq94VTNns+F1z7ALZcO4cpzvsWM2fP57kV3ley6WopPp07l9FOGUFNTw9JYyreOPJqDDxnEDddfx1W/u4Jpn37KLjttz8CBB3PTzX8CYOQD93PANwaw1lprlTj6bMp6qzQf+vIbdxEOLs2PiMoVytoAHwJbRsQ8SX8HbomIRySdBJwAbAUcExEv5Oz3IdAnImbke/6K9l2i7ZZHr/6FWJOZ/erwUodgjbDnrn0YM2Z0QTNh2w17Rbfj87ul8sFVB4+JiD6FPH+hlKJbYCCwLvBWmjD3YvmugfHA0cAISds2fXhmVkoCpPyWLCtFcv02yU2qnhHRE9gUGCBp2Wj3iPg3cAbwiKQeJYjRzEqmPEYLNOlDBGkCPRD4bm1ZRCyQ9DxwaG7diHhY0vrAY5L2Totz+1zfjIgTmyJuM2taGc+beSlqcl2xvzUiFgJfGdQXEbkDOG/PKb8NuC1d7Vn4CM0scwQVGR8JkA8/oWVmmSKS5JrPktfxpFaSXpf0cLq+qaSXJU2QNCK9yY6ktun6xHR7z5xj/Dwtf0/Sgfmc18nVzDKnwDe0fkgyvLPWb4GrI6IXMBs4NS0/FZgdEVsAV6f1kLQNcCywLckN+RsltWropE6uZpY5hbqhJakbcAjwp3RdwP7A/6VV7gAOSz8PTtdJtx+Q1h8M3BsRiyPiP8BEoG9D53ZyNbNsybPVmubWzpJG5yxDVzjaNcBPgdonPdcD5uQ8oDQF6Jp+7krySD7p9rlp/WXlK9mnTp5y0MwyRagxk2XPqOshAkmDgOkRMUZSv2WH/6poYFt9+9TJydXMMqdAQ7H2BL4p6WBgTWAdkpZsB0mt09ZpN+CTtP4UoDswRVJrkoedZuWU18rdp07uFjCzzClEn2tE/DwiuqUPKx0LPB0RxwPPAEem1YaQzHECMDJdJ93+dDoj30jg2HQ0waZAL+CVhq7BLVczy5biP9r6M+BeSZcCrwO3pOW3AHdKmkjSYj0WICLGSboPeAdYApwZETVfPezynFzNLFOSuQUKm10j4lng2fTzB6zkbn9EfA4cVcf+lwGXNeacTq5mljl+/NXMrAjK4fFXJ1czyxaVx2TZTq5mlim187k2d06uZpYx2Z+rNR9OrmaWOWWQW51czSx73HI1Myswlclk2U6uZpY5brmamRVBGeRWJ1czyx63XM3MCq34E7c0CSdXM8uUZLLs5p9dnVzNLHMqyqDp6uRqZplTBrnVydXMskXlPnGLpHXq2zEiPit8OGZmUAZdrvW2XMfx1Tcf1q4H0KOIcZlZC1bWLdeI6F7XNjOzYhHlcUMrr7e/SjpW0vnp526Sdi5uWGbWklUovyXLGkyukoYD+wHfSYsWAr8vZlBm1oLl+VrtrHcd5DNaYI+I2EnS6wARMUtSmyLHZWYtWMbzZl7ySa7VkipIbmIhaT1gaVGjMrMWqyX1ud4A/A1YX9LFwPPAb4salZm1aFJ+S5Y12HKNiD9LGgP0T4uOioi3ixuWmbVULW2y7FZANUnXQF4jDMzMVlWL6BaQdAFwD7Ax0A34i6SfFzswM2u5lOeSZfm0XE8Ado6IhQCSLgPGAL8pZmBm1nJlfZhVPvJJrh+tUK818EFxwjGzli4ZLVDqKFZffRO3XE3Sx7oQGCfp8XR9AMmIATOzwlP5T5ZdOyJgHPBITvlLxQvHzKzMuwUi4pamDMTMDFpAt0AtSZsDlwHbAGvWlkfE14oYl5m1YOXQcs1nzOrtwG0k/6AcBNwH3FvEmMyshSvEUCxJa0p6RdIbksalT5giaVNJL0uaIGlE7Vwpktqm6xPT7T1zjvXztPw9SQfmcw35JNf2EfE4QERMiogLSWbJMjMrOCl5iCCfpQGLgf0jYgegNzBQ0m4kj+9fHRG9gNnAqWn9U4HZEbEFcHVaD0nbAMcC2wIDgRsltWro5Pkk18VK2uiTJJ0h6VCgSx77mZmtkooK5bXUJxLz09U10iWA/YH/S8vvAA5LPw9O10m3H5DmvsHAvRGxOCL+A0wE+jZ4DXlc54+BSuAHwJ7A6cApeexnZrZKGjFxS2dJo3OWocsfR60kjQWmA6OAScCciFiSVpkCdE0/dwUmA6Tb5wLr5ZavZJ865TNxy8vpx3l8OWG2mVlRiLy+8teaERF96toYETVAb0kdgPuBrVdWbdmpV76trvJ61fcQwf31HSAijmjo4GZmjVaE6QQjYo6kZ4HdgA6SWqet027AJ2m1KUB3YIqk1sC6wKyc8lq5+9Spvpbr8EZfQcbsuHUPXni52V9Gi9LxkN+VOgRrhMUTpxXluIUYiiVpfaA6TaztSKZN/S3wDHAkyainIcCD6S4j0/UX0+1PR0RIGkkyYdVVJBNY9QJeaej89T1E8NQqX5WZ2Woo0LymGwF3pHf2K4D7IuJhSe8A90q6FHgdqH1g6hbgTkkTSVqsxwJExDhJ9wHvAEuAM9PuhnrlO5+rmVmTENCqAI9oRcSbwI4rKf+Aldztj4jPgaPqONZlJA9T5c3J1cwyp0U8/lpLUtuIWFzMYMzMkmFWzT+75vMmgr6S3gImpOs7SLq+6JGZWYtVofyWLMun3/g6YBAwEyAi3sCPv5pZEbWIt78CFRHx0QrN9AbvlJmZrQoBrbOeOfOQT3KdLKkvEOmQhu8D7xc3LDNrycogt+aVXL9H0jXQA5gGPJmWmZkVnPKb8Srz8plbYDrpYFozs6ZQBrk1rzcR/JGVzDEQEUNXUt3MbLVlfSRAPvLpFngy5/OawOEsP/2WmVnBJO/Qav7ZNZ9ugRG565LuJJkX0cys8AStCjS5QCmtyuOvmwKbFDoQM7NaavANWdmXT5/rbL7sc60gmS3mvGIGZWYtV4t4tXb6/pgdgP+mRUsjosEZuM3MVkc5JNd6ezbSRHp/RNSkixOrmRWdpLyWLMun2/gVSTsVPRIzM77sFmjuE7fU9w6t2nfM7AWcLmkSsIDk2iMinHDNrPBUmMmyS62+PtdXgJ348p3eZmZF1xJuaAkgIiY1USxmZkD5P/66vqSz69oYEVcVIR4za/FERZmPc20FVEIZXKWZNRui/FuuUyPiV00WiZkZgKB1GXS6NtjnambWlFpCy/WAJovCzCxHWc+KFRGzmjIQM7NaZZBbV2lWLDOzohH5PTqadU6uZpYtIvPzBuTDydXMMkVAKydXM7PCa/6p1cnVzDKoDBquTq5mljXZn6s1H06uZpYp5TJaoByuwczKTCHeRCCpu6RnJL0raZykH6blnSSNkjQh/dkxLZek6yRNlPRm7ksCJA1J60+QNCSfa3ByNbNsUfKEVj5LA5YAP4mIrYHdgDMlbUPygtWnIqIX8BRfvnD1IKBXugwFboIkGQPDgF2BvsCw2oRcHydXM8uU2m6BfJb6RMTUiHgt/TwPeBfoCgwG7kir3cGXLwQYDPw5Ei8BHSRtBBwIjIqIWRExGxgFDGzoOtznamaZU+gbWpJ6AjsCLwMbRMRUSBKwpC5pta7A5JzdpqRldZXXy8nVzDKnEam1s6TROes3R8TNyx1LqgT+BvwoIj6rJ3GvbEPUU14vJ1czy5xGNFxnRESfuo+jNUgS690R8fe0eJqkjdJW60bA9LR8CtA9Z/duwCdpeb8Vyp9tKDD3uZpZptQ+/prPUu9xkibqLcC7K7yWaiRQe8d/CPBgTvmJ6aiB3YC5affB48AASR3TG1kD0rJ6ueVqZhkjVJgHYPcEvgO8JWlsWnY+cDlwn6RTgY+Bo9JtjwIHAxOBhcDJkEy/KukS4NW03q/ymZLVydXMMqcQ97Mi4nnq7r79yssAIiKAM+s41q3ArY05v5OrmWVKMhTLj7+amRWWPHGLmVlROLlawX3++ef0328fvli8mCU1Szj8iCP5xbCLuemG4Qy//ho+mDSJyVOr6Ny5MwBz587llCEnMPnjj1lSs4Qf/fgcTjzp5BJfRcsw/o7TmLfoC2qWBktqlrLX9++m49prcuf5g9hkg3X4aNpnnHDZQ8yZvxiAvbfvxpVn7McarSuYOXcRA869b9mxKirEC9efwCcz5/GtXz5QqkvKBE+WbUXRtm1bHhv1NJWVlVRXV7P/vnsx4MCD2H2PPTn4kEEM6N9vufp/uOkGttp6G/72wENUVVWxw7Zbcuxxx9OmTZvSXEALM/Cnf2XmZ4uWrZ9zdF+eff1j/ve+Vzjn6L6cc0xfLrzlX6y7VluuPas/gy/4G5Or5rH+uu2WO85Zh+3Ee5NnsnZ7/96AQo0WKCmPc80YSVRWVgJQXV3NkupqJNF7xx3ZpGfPldafP28eEcGC+fPp2KkTrVv738xSGbT75tz15DgA7npyHIfuvgUAx+y3FQ++MIHJVfMAqJr7ZULu2rmSgX035bZ/vNX0AWeUlN+SZU6uGVRTU8OuO/emx8Zd2L//N+i766511j3jf85i/Ph32azHxvTZ8ev871XXUlHhX2tTCOChX3+LF4afwCkHfR2ALh3b8+msBQB8OmsB63doD0Cvbh3pULkmj19xNC8MP4Hj+m+z7DhXnrEfF/zpOZZGg09UthjK878sK0oTR9J6JFN5AWwI1ABV6foOwBvpuf8DfCci5kjqB5wTEYNyjnM78HBE/J+kZ4GNgNp/8idGxJHFiL/UWrVqxctjxjJnzhyOOfJwxr39Nttut91K64564nG236E3j416mg8mTeKQg77BnnvtzTrrrNPEUbc8+//4HqbOWsD667bj4cuP5L3JdY8rb92qgp16deGgn/2Vdm3X4Nlrvs0r706lV7eOTJ+zkNcnTmfv7bs1YfTZJaAi23kzL0Vp4kTEzIjoHRG9gd8DV+esL0g/bwfMoo5Bu3U4vvY45ZpYc3Xo0IF99u3HE088VmedO++4jcGHH4EkNt9iC3r23JT3xo9vwihbrqlpC7Vq7iJGvjCRXbbaiOmzF7Jhp7UA2LDTWlTNWQjAf6vm88ToD1m4eAkzP1vE829NYfvN1mf3bTZm0G6bM/6O0/jzzwfRb4ce3PrTg0p2TdmQb7s12xm41N8fXySPqbtakqqqKubMmQPAokWLePqpJ9lyy63qrN+9ew+efTr5kjBt2jTef/89Nt1ssyaJtSVr37Y1le3WWPa5/849GffhDB55aRIn9N8WgBP6b8vDL04C4KEXJ7Lndl1pVSHatW3NLlttxPiPZ/LL255nixNuZqshf+LE3zzMs298zClX/KNk15UJSlqu+SxZVrI7H5JakTyCdksjdrtbUm23wKiIOLfwkZXWp1OncvopQ6ipqWFpLOVbRx7NwYcM4obrr+Oq313BtE8/ZZedtmfgwIO56eY/cd4Fv2DoqSfRp/fXCYLLfv3bZcO0rHi6dFyLEcO+CSRf+Uc8M55Roz9kzHufctcFgxgycDsmT/+M4y97GID3Js9i1OgPefX3Q1gawe2PvcU7H80s5SVkVtItkPHMmQdFkTvRJV0EzI+I/03Xa4C3gJ7AGGBARNRI2hc4dyV9rg9FxN/SPtdzImI09ZA0lOQVDXTv0WPn9yd9VPBrsuLpeMjvSh2CNcLil65h6dzJBc2EW399x7jt/mfyqrt7r45j6ptysJRK0S2wKO173QRow5d9rjOBFd9L0wmY0ZiDR8TNEdEnIvqs33n91Q7WzEpAeS4ZVrI+14iYC/wAOCed0HYCsLGkrQEkbUIysmBs3Ucxs3JUDje0SjraPCJel/QGcGxE3CnpBOA2SWsC1cBpaRKuldvnOiMi+jd1zGZWfFm/WZWPoifXiLhohfXKFdYPzfn8AskrcFd2nH5FCM/MssjJ1cyssJLu1OafXZ1czSxbmsG8AflwcjWzzCmD3OrkamYZVAbZ1cnVzDJGZfGElpOrmWVKM3g+IC9OrmaWPWWQXZ1czSxzPBTLzKwIyqDL1cnVzLKnDHKrk6uZZYySF282d06uZpYpwt0CZmZFUQa51cnVzDKoDLKrk6uZZY6HYpmZFYEnyzYzK4YySK4le4eWmdnK1E6WXYh3aEm6VdJ0SW/nlHWSNErShPRnx7Rckq6TNFHSm5J2ytlnSFp/gqQh+VyHk6uZZUs6WXY+Sx5uBwauUHYe8FRE9AKeStcBDgJ6pctQ4CZIkjEwDNgV6AsMq03I9XFyNbPMKdSbtSPiOWDWCsWDgTvSz3cAh+WU/zkSLwEdJG0EHAiMiohZETEbGMVXE/ZXuM/VzLIn/z7XzpJG56zfHBE3N7DPBhExFSAipkrqkpZ3BSbn1JuSltVVXi8nVzPLmEZNlj0jIvoU7MRfFfWU18vdAmaWKfl2CazGgIJp6dd90p/T0/IpQPecet2AT+opr5eTq5llT3Gz60ig9o7/EODBnPIT01EDuwFz0+6Dx4EBkjqmN7IGpGX1creAmWVOoZ7QknQP0I+kb3YKyV3/y4H7JJ0KfAwclVZ/FDgYmAgsBE4GiIhZki4BXk3r/SoiVrxJ9hVOrmaWOYWaFSsivl3HpgNWUjeAM+s4zq3ArY05t5OrmWVOGTyg5eRqZhnjybLNzArPk2WbmRVJGeRWJ1czyx63XM3MisCTZZuZFYFbrmZmBdaI6QQzzcnVzDLH3QJmZsXQ/HOrk6uZZU8Z5FYnVzPLHve5mpkVmBo3WXZmeT5XM7MicMvVzDKnDBquTq5mlj0eimVmVmh+iMDMrPBW8+WDmeHkamaZ48myzcyKoAxyq5OrmWVPGeRWJ1czy6AyyK5OrmaWOeUwFEvJq7rLk6Qq4KNSx1EEnYEZpQ7CGqVcf2ebRMT6hTygpMdI/rzyMSMiBhby/IVS1sm1XEkaHRF9Sh2H5c+/s5bHcwuYmRWBk6uZWRE4uTZPN5c6AGs0/85aGPe5mpkVgVuuZmZF4ORqZlYETq5lQNJ6pY7BGkfSXpJ6lzoOKx4n12ZO0gDgGkkdVQ5TCbUAkg4EhgNtSh2LFY+TazOWJtYrgVsiYjZ+nDnz0sR6OYr8o00AAAe7SURBVPDDiHhF0vqSNi91XFZ4Tq7NlKSBJIn1uxHxrKTuwPmS8n1s0JqYpC1IhmRdFRH/lNQNuA/YsrSRWTE4uTZfuwLtI+IlSesD9wPTI6Icn19v9iS1i4iJwBNAb0l7AyOAv0fEo6WNzorBybWZkbSnpPMj4mLg35JeBB4G/hARN+XU616yIG05krYGLpHUKyJOB9YEbgdGRcT1OfWOlLRLicK0AnNybSYk1f6uBgDrAkTEEOA5oGNE/DGn7vHAdZLWbvJAbWVmAx2BUyVtHhFnAg8CG0naDEDSscCFwGelC9MKycm1+Vg3/fk5UF1bGBE/A56X9AyApG8BPwYuiIh5TR6lfUVEfAqcD1QCZ0jaIiLOBgI4W9K5wE+A4yLivRKGagXk5NoMSNoU+E3aypkGrJ2WtwOIiFOADyRNJflLfGJEvFOqeA0kfV3SpbXrETENuJSkS+A0SR0i4gxgHeA7wBD/zsqLk2vzsCYwHfgusBcwNy1vWzu2NSJOBX4HfMd/SUtHKaA9sK2ki2q3pS3Yy4DuJC1VIuJEoL9/Z+XHE7c0E5K2AwYCZwE9gJHAjsAnJN0E84DDIqK6zoNY0UlqFRE16eedSZLopIj4RU6drUn+ITw6IuaXJlIrNg86zyhJ/Uh+P89FxBcR8bakapIW0dYkd5vfAtYi+Wo53Ym1tCT1B06W9CbwSkQ8I+lq4AeSLslJsL1J/kH8olSxWvG55ZpBktYFHgE2Ba4BaiLiqnTb5sAxwEbAnRHxSskCtWXSJ69+TTJ2tTPQAbiI5L1ZXwd+RtJXPgo4gaRf/O2SBGtNwn2uGRQRc0nGrn4BTAAOlnS7pMNI+l5vIBnec7SkNT2nQGmlifVB4OyIuAL4I8k3io7pt44xwLeBp0mGWh3nxFr+3HLNEEkbAtMiIiStAVwL/C0inpL0F2A3kptZw4CFwGsRMat0EVuaWC8BlgJExG5p+RMkX/3HAh8AfwaWhP/CtRhuuWaEpENIblJ1Th8YEEkrdcd0CNZuwMkk3QRHAK87sZZWzuxWp6VJdbKkVyVdSTLj1f3ALJKbWteSjHO1FsIt1wxIJ2G5ALgsIh6T1CYivkgfYR1D8pfy6Ih4OK3fPiIWljDkFi+dkexO4F/AsIgYl5bfBgwB1oyIL9KySpJ5IKaXKl5rek6uJSapE8lNjyMi4oH0htUvgXMjYrqkocD2EXFWbdItacCGpAOAm4CLgQ2ALsBjEfFsun0EyVjWfT2Co+Vyt0CJpV/tDwV+KWl7kinpXs9p5bwBHCDpa06smfEZcFJE3E0yqqMaODAdPkdEHEPSN/5kySK0knPLNSPSroFHgfMj4vIVBqP/AhgREe+XNEhbjqSKiFgqqRfJI6xrAP+IiOfS7V0j4r8lDdJKxsk1QyR9A7ge2DUi5kpqGxGLSx2XNSxNsMcB65H8Q/iCJHl0QMvlboEMiYhRJDNavSKpkxNr8xERE0geIJgKvJ+WObG2YG65ZpCkwSRjWfuQ/B31L6mZkLSGb2IZOLlmlqRKT+ph1nw5uZqZFYH7XM3MisDJ1cysCJxczcyKwMnVzKwInFzNzIrAybUFk1QjaayktyX9VVL71ThWP0m1s3Z9U9J59dTtIOl/VuEcF0k6J9/yFercLunIRpyrpyRPaG2rzMm1ZVsUEb0jYjuStx6ckbsxfZFpo/8fiYiREXF5PVU6AI1OrmbNiZOr1foXsEXaYntX0o3Aa0B3SQMkvSjptbSFWwnJZDOSxkt6nmQCb9LykyQNTz9vIOl+SW+kyx7A5cDmaav5yrTeuelE029KujjnWBdIek/Sk8CWDV2EpNPT47wh6W8rtMb7S/qXpPclDUrrt5J0Zc65v7u6f5Bm4ORqgKTWwEEkb5OFJIn9OSJ2BBYAFwL9I2InYDRwtqQ1Sd4VdSiwN7BhHYe/DvhnROwA7ASMA84jed1074g4N514uhfQl+TNqDtL2id9NfWxJK8QPwLYJY/L+XtE7JKe713g1JxtPYF9gUOA36fXcCowNyJ2SY9/uqRN8ziPWb38au2WrZ2ksennfwG3ABsDH0XES2n5bsA2wAvpexDbAC8CWwH/SScsQdJdwNCVnGN/4ESAdArFuZI6rlBnQLq8nq5XkiTbtYH7a9+6IGlkHte0naRLSboeKoHHc7bdFxFLgQmSPkivYQCwfU5/7LrpuT29o60WJ9eWbVFE9M4tSBPogtwiYFREfHuFer2BQj07LeA3EfGHFc7xo1U4x+3AYRHxhqSTgH4521Y8VqTn/n5E5CZhJPVs5HnNluNuAWvIS8CekraA5P1dkr4GjAc2TV9LA8mro1fmKeB76b6tJK0DzCNpldZ6HDglpy+3q6QuwHPA4ZLaSVqbpAuiIWsDU5W8Pff4FbYdJakijXkz4L303N9L6yPpa5LWyuM8ZvVyy9XqFRFVaQvwHklt0+ILI+L99P1ej0iaATwPbLeSQ/wQuFnSqUAN8L2IeFHSC+lQp3+k/a5bAy+mLef5wAkR8ZqS91GNBT4i6bpoyC+Al9P6b7F8En8P+CfJe6/OiIjPJf2JpC/2NSUnrwIOy+9Px6xunhXLzKwI3C1gZlYETq5mZkXg5GpmVgROrmZmReDkamZWBE6uZmZF4ORqZlYE/w+NxEJ+kh9ArQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "passive = PassiveAggressiveClassifier()\n",
    "passive.fit(count_train, y_train)\n",
    "pred = passive.predict(count_test)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\sid321axn\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:939: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html.\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.951\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clf = LogisticRegression()\n",
    "clf.fit(count_train, y_train)\n",
    "pred = clf.predict(count_test)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TFIDF Experiment (n_gram=2 (bigram) and max_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfidf_vectorizer = TfidfVectorizer(stop_words='english', max_df=0.8, ngram_range=(1,2))\n",
    "tfidf_train_2 = tfidf_vectorizer.fit_transform(X_train)\n",
    "tfidf_test_2 = tfidf_vectorizer.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.941\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mnb_tf = MultinomialNB()\n",
    "mnb_tf.fit(tfidf_train_2, y_train)\n",
    "pred = mnb_tf.predict(tfidf_test_2)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.959\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pass_tf = PassiveAggressiveClassifier()\n",
    "pass_tf.fit(tfidf_train_2, y_train)\n",
    "pred = pass_tf.predict(tfidf_test_2)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.946\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "log_tf = LogisticRegression()\n",
    "log_tf.fit(tfidf_train_2, y_train)\n",
    "pred = log_tf.predict(tfidf_test_2)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TFIDF Experiment (n_gram=3 (trigram) and max_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfidf_vectorizer_3 = TfidfVectorizer(stop_words='english', max_df=0.8, ngram_range=(1,3))\n",
    "tfidf_train_3 = tfidf_vectorizer_3.fit_transform(X_train)\n",
    "tfidf_test_3 = tfidf_vectorizer_3.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.941\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mnb_tf2 = MultinomialNB()\n",
    "mnb_tf2.fit(tfidf_train_3, y_train)\n",
    "pred = mnb_tf2.predict(tfidf_test_3)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.958\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVcAAAEmCAYAAADWT9N8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3dd5wV1fnH8c93d6kCCgKKFLGgoiaiIhJ7LIgVY+wNFYMaNSbGFEtij0b92WJJNHYTxRiNBAuisXdQFLGBlaaIFBEFKc/vj5nFK7K7d+Fe7uzd79vXvPbOmTMzz7DycO6ZM2cUEZiZWWFVlDoAM7Ny5ORqZlYETq5mZkXg5GpmVgROrmZmReDkamZWBE6uVitJLST9V9IsSf9ajuMcKumRQsZWKpK2lfROqeOwbJPHuZYHSYcApwAbALOB0cAFEfHMch73cOAkYKuIWLDcgWacpAB6RMT4UsdiDZtbrmVA0inAFcCfgNWAbsC1wIACHH5N4N3GkFjzIamq1DFYAxERXhrwAqwMfAnsX0udZiTJd3K6XAE0S7ftAEwEfg1MBaYAR6XbzgG+Aean5xgEnA3ckXPs7kAAVen6kcD7JK3nD4BDc8qfydlvK+BlYFb6c6ucbU8A5wHPpsd5BGhfw7VVx//bnPj3AXYH3gWmA6fn1O8DPA/MTOteDTRNtz2VXsuc9HoPzDn+74BPgNury9J91knPsVm6vgYwDdih1P9veCnt4pZrw/cjoDlwXy11zgD6Ar2ATUgSzJk521cnSdKdSRLoNZLaRsRZJK3hIRHRKiJurC0QSSsBVwG7RURrkgQ6ein12gEPpHVXBS4DHpC0ak61Q4CjgI5AU+DUWk69OsmfQWfgj8ANwGHA5sC2wB8lrZ3WXQj8CmhP8me3E/BzgIjYLq2zSXq9Q3KO346kFT8498QR8R5J4v2HpJbAzcAtEfFELfFaI+Dk2vCtCkyL2r+2HwqcGxFTI+Izkhbp4Tnb56fb50fEgySttvWXMZ5FwMaSWkTElIgYu5Q6ewDjIuL2iFgQEXcCbwN75dS5OSLejYivgbtJ/mGoyXyS/uX5wF0kifPKiJidnn8s8EOAiBgVES+k5/0Q+BuwfR7XdFZEzEvj+Y6IuAEYB7wIdCL5x8waOSfXhu9zoH0dfYFrAB/lrH+Uli0+xhLJ+SugVX0DiYg5JF+ljwOmSHpA0gZ5xFMdU+ec9U/qEc/nEbEw/Vyd/D7N2f519f6S1pM0TNInkr4gaZm3r+XYAJ9FxNw66twAbAz8JSLm1VHXGgEn14bveWAuST9jTSaTfKWt1i0tWxZzgJY566vnboyI4RGxC0kL7m2SpFNXPNUxTVrGmOrjOpK4ekREG+B0QHXsU+uQGkmtSPqxbwTOTrs9rJFzcm3gImIWST/jNZL2kdRSUhNJu0m6OK12J3CmpA6S2qf171jGU44GtpPUTdLKwGnVGyStJmnvtO91Hkn3wsKlHONBYD1Jh0iqknQgsCEwbBljqo/WwBfAl2mr+vgltn8KrP29vWp3JTAqIo4h6Uv+63JHaQ2ek2sZiIjLSMa4ngl8BkwATgT+k1Y5HxgJvA6MAV5Jy5blXCOAIemxRvHdhFhBMupgMskd9O1JbxYtcYzPgT3Tup+T3OnfMyKmLUtM9XQqyc2y2SSt6iFLbD8buFXSTEkH1HUwSQOA/iRdIZD8HjaTdGjBIrYGyQ8RmJkVgVuuZmZF4ORqZlYETq5mZkXg5GpmVgRlPQmFqlqEmrYudRhWD5v27FbqEKwePvroQ6ZNm1bXOOF6qWyzZsSC7z0It1Tx9WfDI6J/Ic9fKOWdXJu2ptn6dY6msQx59sWrSx2C1cPWW/Yu+DFjwdd5/72dO/qaup6uK5myTq5m1hAJ1PB7LJ1czSxbBFRUljqK5ebkambZo4J245aEk6uZZYy7BczMisMtVzOzAhNuuZqZFZ7ccjUzKwqPFjAzKzTf0DIzKzzhbgEzs6Jwy9XMrNDcLWBmVngCKn1Dy8ys8NznamZWaO4WMDMrDrdczcyKwC1XM7MCkx9/NTMrDj/+amZWaL6hZWZWHO4WMDMrsDKZz7XhX4GZlZm0WyCfJZ+jSR9KGiNptKSRaVk7SSMkjUt/tk3LJekqSeMlvS5ps5zjDEzrj5M0sK7zOrmaWfZUjxioa8nfjyOiV0T0Ttd/DzwWET2Ax9J1gN2AHukyGLguCUftgLOALYE+wFnVCbkmTq5mlj0Vlfkty24AcGv6+VZgn5zy2yLxArCKpE7ArsCIiJgeETOAEUD/Wi9heaIzMys41atboL2kkTnL4KUcMYBHJI3K2b5aREwBSH92TMs7AxNy9p2YltVUXiPf0DKz7Mn/K/+0nK/6Ndk6IiZL6giMkPR2bWdeSlnUUl4jt1zNLHMk5bXkIyImpz+nAveR9Jl+mn7dJ/05Na0+Eeias3sXYHIt5TVycjWzTEne8lKY5CppJUmtqz8D/YA3gKFA9R3/gcD96eehwBHpqIG+wKy022A40E9S2/RGVr+0rEbuFjCzbJFQRcEeIlgNuC9NxFXAPyPiYUkvA3dLGgR8DOyf1n8Q2B0YD3wFHAUQEdMlnQe8nNY7NyKm13ZiJ1czy5x8v/LXJSLeBzZZSvnnwE5LKQ/ghBqOdRNwU77ndnI1s8wpVHItJSdXM8scJ1czs0ITSx/41MA4uZpZpoj8h1llmZOrmWVORUXDHyXq5GpmmeOWq5lZobnP1cysONxyNTMrMN/QMjMrEidXM7NCE4WcW6BknFzNLHPccjUzKwInVzOzAvMNLTOzYmn4udXJNSvefuAcZs+Zx8JFi1iwcBHbHHoxAMcftD3HHbgdCxYu4uGn3+CMK5MJ0zfusQZXn3kwrVdqzqJFwTaHXcy8bxZw/9U/Z/UObaiqrOTZV9/jlxcOYdGiWl/1YwVw7DFH89CDw+jQsSOjRr8BwGujR3PSCccxb+5cqqqquOIv17JFnz5EBL/+1ckMf/hBWrZoyfU33sKmm21W4ivIEPnxVyuw/oOv5POZcxavb9e7B3vu8AO2OOBCvpm/gA5tWwFQWVnBTecPZNAfbmPMu5Not/JKzF+wEIDDfncTs+fMBeDOS4/hp7tsxr+Gj1rxF9PIHD7wSI77+Ykcc/QRi8vOOO23nPGHs9i1/248/NCDnHHab3nksScY/vBDvDd+HG+8NY6XXnyRX5x4PE8/92IJo8+ecugWaPj/PJSxwftvy6U3j+Cb+QsA+GzGlwDs/KMNeGPcJMa8OwmA6bPmLG6dVifWqqoKmlRVkkysbsW2zbbb0a5du++USeKLL74AYNasWXRaYw0Ahg29n0MOOwJJbNm3L7NmzWTKlCkrPOZMU55LhrnlmhERwX+vPZGI4MZ/P8tN9z7Lumt2ZOtN1+GcE/Zi7jfzOe2y+xj15sf06NaRCBh6zQm0b9uKe4aP4rJbH118rKHXnEDvjdfkkWff5N5HXy3hVTVul/zfFey1x66c9rtTWbRoEY8/9RwAkydPokuXb18k2rlzFyZPmkSnTp1KFWrmuOVaB0kLJY3OWbrnbLtS0iRJFTllR0q6Ov1cIelWSTelb2L8UNKYnGNdVczYV7Qdj7qcrQ75M/uceC3HHrgtW2+2DlWVFbRt05LtjriU0y//D3dcfDQAVZWVbLXp2hx1xi3sdPRl7L3jJuzQZ73Fx9r7hGtYa5fTada0ih22WL9Ul9ToXf+367j40ssZ/8EELr70co4fPAhgqd8myiGZFEq+b37N+p9ZsbsFvo6IXjnLh5AkTuAnwARguyV3UvKn9legCXBMfPt/449zjvWLIse+Qk35bBaQfPUf+r/X2WKj7kz6dCb/eew1AEaO/YhFi4L2bVsxaepMnh41ns9nzuHrufN5+JmxbLpB1+8cb943Cxj25Bj22uEHK/xaLPGP229ln5/sC8BP99ufkS+/BCQt1YkTJyyuN2nSxMVdBpZwcl12PyZ5d/h1wMFL2X4lsCpwREQsWpGBlULL5k1p1bLZ4s87/2gDxr43mf8+8friFum63TrStEkV02Z8yYjn3mTjHp1p0bwJlZUVbLv5urz1/ies1KIpq7dvAyQ3vfpvvSHvfPhpya6rseu0xho8/dSTADzx+P9Yd90eAOyx1978847biAhefOEF2rRZ2V0CS1CF8lqyrNh9ri0kjU4/fxARP0k/HwzcCdwP/ElSk4iYn247BHgL2CEiFixxvMclLUw/3xoRly95QkmDgcEANGlVuCspoo6rtmbIZT8Dkq/8Qx4ayYjn3qJJVSV/O/tQRv7rdL6Zv5Bj/ng7ADNnf81Vd/yPZ+74LRHB8GfG8vAzY+nYrjX3XHEsTZtUUVlZwZMvv8sN9zxTyktrNI447GCefvIJpk2bxjrdu/CHP57DNdfdwG9OOZkFCxbQrHlzrr7uegD677Y7wx96kI02WJeWLVryt7/fXOLosyfrrdJ8qJh3kyV9GRGtlihrCnwIrB8RsyXdC9wYEQ9IOhI4DNgAODAins3Z70Ogd0RMy/f8FS07RrP1D1j+C7EVZsbLV5c6BKuHrbfszahRIwuaCZut3iO6HJrfLZX3L9t9VET0LuT5C6UU3QL9gZWBMWnC3Ibvdg28DRwADJG00YoPz8xKSYCU35JlpUiuB5PcpOoeEd2BtYB+klpWV4iI54DjgAckdStBjGZWMuUxWmCFjnNNE+iuwLHVZRExR9IzwF65dSNimKQOwMOStk2Lc/tcX4+IIzCzspPxvJmXoibXJftbI+IroN1S6u2bs3pLTvnNQHVvf/fCR2hmmSOoyPhIgHz4CS0zyxTh5GpmVhTl0C3giVvMLHMKeUNLUqWkVyUNS9fXkvSipHGShqTDQ5HULF0fn27vnnOM09LydyTtms95nVzNLFvyHIZVj9btySQPJlX7M3B5RPQAZgCD0vJBwIyIWBe4PK2HpA2Bg4CNSIaSXiupsq6TOrmaWaYIUVFRkddS57GkLsAewN/TdQE7AvekVW4F9kk/D0jXSbfvlNYfANwVEfMi4gNgPNCnrnM7uZpZ5tSj5dpe0sicZfASh7oC+C1QPUfJqsDMnEfrJwKd08+dSSaTIt0+K62/uHwp+9TIN7TMLHPq8YDAtJoef5W0JzA1IkZJ2qG6eClVo45tte1TIydXM8uWwj3aujWwt6TdgeZAG5KW7CqSqtLWaRdgclp/ItAVmCipiuQx/ek55dVy96mRuwXMLFOSuQWWf7RARJwWEV3Sx+wPAv4XEYcCjwP7pdUGkszOBzA0XSfd/r90LumhwEHpaIK1gB7AS3Vdh1uuZpY5RR7n+jvgLknnA68CN6blNwK3SxpP0mI9CCAixkq6G3gTWACcEBELv3/Y73JyNbPMKfQTWhHxBPBE+vl9lnK3PyLmAvvXsP8FwAX1OaeTq5lli8pjsmwnVzPLlOr5XBs6J1czy5jsz9WaDydXM8ucMsitTq5mlj1uuZqZFZg8WbaZWXG45WpmVgRlkFudXM0se9xyNTMrtMJN3FJSTq5mlinJZNkNP7s6uZpZ5lSUQdPVydXMMqcMcquTq5lli8p94hZJbWrbMSK+KHw4ZmZQBl2utbZcx/L998dUrwfQrYhxmVkjVtYt14joWtM2M7NiEeVxQyuvd2hJOkjS6ennLpI2L25YZtaYVSi/JcvqTK6SrgZ+DByeFn0F/LWYQZlZI5bnywmz3nWQz2iBrSJiM0mvAkTEdElNixyXmTViGc+becknuc6XVEFyEwtJqwKLihqVmTVajanP9Rrg30AHSecAzwB/LmpUZtaoSfktWVZnyzUibpM0Ctg5Ldo/It4oblhm1lg1tsmyK4H5JF0DeY0wMDNbVo2iW0DSGcCdwBpAF+Cfkk4rdmBm1ngpzyXL8mm5HgZsHhFfAUi6ABgFXFjMwMys8cr6MKt85JNcP1qiXhXwfnHCMbPGLhktUOooll9tE7dcTtLH+hUwVtLwdL0fyYgBM7PCU/lPll09ImAs8EBO+QvFC8fMrMy7BSLixhUZiJkZNIJugWqS1gEuADYEmleXR8R6RYzLzBqxcmi55jNm9RbgZpJ/UHYD7gbuKmJMZtbIFWIolqTmkl6S9JqksekTpkhaS9KLksZJGlI9V4qkZun6+HR795xjnZaWvyNp13yuIZ/k2jIihgNExHsRcSbJLFlmZgUnJQ8R5LPUYR6wY0RsAvQC+kvqS/L4/uUR0QOYAQxK6w8CZkTEusDlaT0kbQgcBGwE9AeulVRZ18nzSa7zlLTR35N0nKS9gI557GdmtkwqKpTXUptIfJmuNkmXAHYE7knLbwX2ST8PSNdJt++U5r4BwF0RMS8iPgDGA33qvIY8rvNXQCvgF8DWwM+Ao/PYz8xsmdRj4pb2kkbmLIO/exxVShoNTAVGAO8BMyNiQVplItA5/dwZmACQbp8FrJpbvpR9apTPxC0vph9n8+2E2WZmRSHy+spfbVpE9K5pY0QsBHpJWgW4D+i5tGqLT730bTWV16q2hwjuq+0AEbFvXQc3M6u3IkwnGBEzJT0B9AVWkVSVtk67AJPTahOBrsBESVXAysD0nPJqufvUqLaW69X1voKM6dWzG089d1Wpw7B6aDvgL6UOweph3vipRTluIYZiSeoAzE8TawuSaVP/DDwO7Ecy6mkgcH+6y9B0/fl0+/8iIiQNJZmw6jKSCax6AC/Vdf7aHiJ4bJmvysxsORRoXtNOwK3pnf0K4O6IGCbpTeAuSecDrwLVD0zdCNwuaTxJi/UggIgYK+lu4E1gAXBC2t1Qq3znczUzWyEEVBbgEa2IeB3YdCnl77OUu/0RMRfYv4ZjXUDyMFXenFzNLHMaxeOv1SQ1i4h5xQzGzCwZZtXws2s+byLoI2kMMC5d30SS7zqYWdFUKL8ly/LpN74K2BP4HCAiXsOPv5pZETWKt78CFRHx0RLN9DrvlJmZLQsBVVnPnHnIJ7lOkNQHiHRIw0nAu8UNy8waszLIrXkl1+NJuga6AZ8Cj6ZlZmYFp/xmvMq8fOYWmEo6mNbMbEUog9ya15sIbmApcwxExOClVDczW25ZHwmQj3y6BR7N+dwc+AnfnX7LzKxgkndoNfzsmk+3wJDcdUm3k8yLaGZWeILKAk0uUErL8vjrWsCahQ7EzKya6nxDVvbl0+c6g2/7XCtIZov5fTGDMrPGq1G8Wjt9f8wmwKS0aFFE1DkDt5nZ8iiH5Fprz0aaSO+LiIXp4sRqZkUnKa8ly/LpNn5J0mZFj8TMjG+7BRr6xC21vUOr+h0z2wA/k/QeMIfk2iMinHDNrPBUmMmyS622PteXgM349p3eZmZF1xhuaAkgIt5bQbGYmQHl//hrB0mn1LQxIi4rQjxm1uiJijIf51oJtIIyuEozazBE+bdcp0TEuSssEjMzAEFVGXS61tnnama2IjWGlutOKywKM7McZT0rVkRMX5GBmJlVK4PcukyzYpmZFY3I79HRrHNyNbNsEZmfNyAfTq5mlikCKp1czcwKr+GnVidXM8ugMmi4OrmaWdZkf67WfJTDTTkzKyPVowXyWWo9jtRV0uOS3pI0VtLJaXk7SSMkjUt/tk3LJekqSeMlvZ47j7WkgWn9cZIG5nMdTq5mljkFehPBAuDXEdET6AucIGlDkncAPhYRPYDH+PadgLsBPdJlMHBdGks74CxgS6APcFZ1Qq6Nk6uZZYuSJ7TyWWoTEVMi4pX082zgLaAzMAC4Na12K9/OWT0AuC0SLwCrSOoE7AqMiIjpETEDGAH0r+sy3OdqZplSz4cI2ksambN+fURc/71jSt2BTYEXgdUiYgokCVhSx7RaZ2BCzm4T07Kaymvl5GpmmVOPG1rTIqJ3HcdqBfwb+GVEfFHLsZe2IWopr5W7Bcwsc5TnUudxpCYkifUfEXFvWvxp+nWf9OfUtHwi0DVn9y7A5FrKa+XkamaZI+W31H4MCbgReGuJN6cMBarv+A8E7s8pPyIdNdAXmJV2HwwH+klqm97I6peW1crdAmaWKQV8/HVr4HBgjKTRadnpwEXA3ZIGAR8D+6fbHgR2B8YDXwFHQTJDoKTzgJfTeufmM2ugk6uZZYxQAR6AjYhnqLn34HvzVUdEACfUcKybgJvqc34nVzPLnDJ4QMvJ1cyyJRmK1fCzq5OrmWVLHjerGgInVzPLHCdXK7iJEyYweNCRfPrpJ1RUVHDUoJ/x8xN/wZjXX+Pkk37OnC+/pNuaa3LjLXfQpk0bRr78Er844TgAIoLTzvwjew/4SWkvopF4+6aBzP76GxYuChYsXMQ2v7ybtq2acfvv+7NmxzZ8NPULDrvoYWZ+OY82LZty06n96NqhNVWV4op7X+X2R98C4PyjtqJ/7+4AXHTXy9zz9LgSXlXpebJsK4qqqir+9OdL6LXpZsyePZttf7QFO+60MyceP5gLLryYbbbbnttuuYkrL7uUP5x9LhtutDFPPfcSVVVVfDJlCj/qsym777EXVVX+1a4I/U+7j8+/mLt4/dT9N+eJ1yZy6b9Gcer+m3Pq/ptz5s3PceyeP+TtCdPZ79xhtG/TnNeuP5y7nniHnTbtRq91OrDlSXfSrEklj/x5X4aP/JDZX88v4VWVXiFGC5SaHyLImNU7daLXpslMZ61bt2b9DTZg8qRJjHv3HbbedjsAdtxpF+7/T/KwScuWLRcn0rlz55bFPJgN2Z591+aOtEV6x6NvsVfftYHkW0WrFk0AWKlFU2bMnsuChYvo2bUtT4+ZxMJFwVfzFjDmg2n023zNksWfFYV4iKDUnFwz7KMPP+T10aPp3WdLem60MQ8MGwrAfffew6SJ384j8fJLL7LFpj+gb+9NuOIv17rVuoJEBP89bwDPXnkgR/ffCICOq7TkkxlfAfDJjK/osEoLAP467HU26NqO928/mpHXHMyp1z9NBLz+wTR27b0mLZpVsWqb5mz/wy506dC6ZNeUFcrzvywryt9CSauSzJMIsDqwEPgsXd8EeC099wfA4RExU9IOwKkRsWfOcW4BhkXEPZKeADoBX6ebx0fEfsWIPwu+/PJLDjt4fy669DLatGnDtX/7O7895WT+/Kfz2X2PvWjStOniulv02ZKXXx3D22+/xXHHHEW/XXejefPmJYy+cdjxN/9myvQ5dFi5BcPO34d3Jsyose4um3Xj9fc/o/9p97F2p5V54PwBPPvGJB57dQKbr7caj1+6H9Nmfc2Lb33CgoWLVuBVZI+AimznzbwUpeUaEZ9HRK+I6AX8Fbg8Z31O+nljYDo1PBFRg0Orj1POiXX+/PkcdtB+HHDQIQzYZ18A1l9/A+5/YDhPP/8y+x14EGuvvc739ttgg560bLkSb459Y0WH3ChNmT4HgM9mfc3Q599ji/VXY+rMr1i9bUsAVm/bks9mJm2Bw3fZkPufex+A96fM4sNPv2D9ru0AuHjISPqedBd7nnk/EoyfPLMEV5Ml+bZbs52BS90t8Dx5zIvYmEQEJxx7DOtv0JOTTv7V4vLPpiYT9yxatIhLLryAo48ZDMCHH3zAggULAPj4o48YN+4duq3ZfYXH3di0bFa1uA+1ZbMqdt6sG2M/+pwHXvyAw3buCcBhO/dk2AtJQp0wdTY7bNIFgI6rtGC9zm354JNZVFSIdq2Tbxkbd1+Vjbu359FXPi7BFWWIkpZrPkuWlaxzTlIlyfO9N9Zjt39Iqu4WGBERvyl8ZKX1/HPPcuc/72CjjX/AVn2SG1tnnXs+740fz/V/vRaAvff5CYcPPCqt/wyXXXoxTZo0oaKigsuuvJr27duXLP7GomPblgw5Yw8AqirFkCffZcSojxn17lTu+H1/Bu6yIRM+m82hFz4EJEOsrv/Vzrx8zcEIccYtz/H5F3Np1qSSRy/+KQCzv/qGo//vERYuqnOq0LKWdAtkPHPmQclcBUU8gXQ28GVEXJquLwTGAN2BUUC/iFgoaXvgN0vpc/1vRPw77XM9NSJGUgtJg0nef0PXrt02f3PcBwW/JiueDvteU+oQrB7mPXMJi2Z9XNBM2PMHm8bN9z2eV90f9Wg7qq7JskulFN0CX6d9r2sCTfm2z/VzYMmXfrUDptXn4BFxfUT0joje7Tt0WO5gzawECjVbdgmVrM81ImYBvwBOTWcLHwesIakngKQ1SUYWjK75KGZWjsrhhlZJB0RGxKuSXgMOiojbJR0G3CypOTAfOCZNwtVy+1ynRcTOKzpmMyu+rN+sykfRk2tEnL3Eeqsl1vfK+fwsyfvFl3acHYoQnpllkZOrmVlhJd2pDT+7OrmaWbY0gHkD8uHkamaZUwa51cnVzDKoDLKrk6uZZYzK4gktJ1czy5QG8HxAXpxczSx7yiC7OrmaWeZ4KJaZWRGUQZerk6uZZU8Z5FYnVzPLGFEWL9p0cjWzTBHuFjAzK4oyyK0lf4eWmdn3FWiybEk3SZoq6Y2csnaSRkgal/5sm5ZL0lWSxkt6XdJmOfsMTOuPkzQwn0twcjWzzCngZNm3AP2XKPs98FhE9AAeS9cBdgN6pMtg4DpIkjFwFrAl0Ac4qzoh18bJ1cwyp1Bvf42Ip4DpSxQPAG5NP98K7JNTflskXgBWkdQJ2JXkhajTI2IGMILvJ+zvcZ+rmWVP/p2u7SXlvrT0+oi4vo59VouIKQARMUVSx7S8MzAhp97EtKym8lo5uZpZptRzsuxpBXz769JOGrWU18rdAmaWLelk2fksy+jT9Os+6c+paflEoGtOvS7A5FrKa+XkamaZU+Q3aw8Fqu/4DwTuzyk/Ih010BeYlXYfDAf6SWqb3sjql5bVyt0CZpY9BRroKulOYAeSvtmJJHf9LwLuljQI+BjYP63+ILA7MB74CjgKICKmSzoPeDmtd25ELHmT7HucXM0sYwo3WXZEHFzDpp2WUjeAE2o4zk3ATfU5t5OrmWWKJ8s2MyuWMsiuTq5mljmeLNvMrAg8K5aZWRGUQW51cjWzjPFk2WZmhefJss3MiqQMcquTq5llj1uuZmZF4KFYZmZF4JarmVmBLed0gpnh5GpmmeNuATOzYmj4udXJ1cyypwxyq5OrmWWP+1zNzApMBZwsu5T8Di0zsyJwy9XMMqcMGq5OrmaWPR6KZWZWaH6IwMys8PyCQjOzIvFk2SDFmkEAAAiHSURBVGZmRVAGudXJ1cyypwxyq5OrmWVQGWRXJ1czy5xyGIqliCh1DEUj6TPgo1LHUQTtgWmlDsLqpVx/Z2tGRIdCHlDSwyR/XvmYFhH9C3n+Qinr5FquJI2MiN6ljsPy599Z4+O5BczMisDJ1cysCJxcG6brSx2A1Zt/Z42M+1zNzIrALVczsyJwcjUzKwIn1zIgadVSx2D1I2kbSb1KHYcVj5NrAyepH3CFpLYqh6mEGgFJuwJXA01LHYsVj5NrA5Ym1kuAGyNiBn6cOfPSxHoRcHJEvCSpg6R1Sh2XFZ6TawMlqT9JYj02Ip6Q1BU4XVK+jw3aCiZpXZIhWZdFxJOSugB3A+uXNjIrBifXhmtLoGVEvCCpA3AfMDUiyvH59QZPUouIGA88AvSStC0wBLg3Ih4sbXRWDE6uDYykrSWdHhHnAM9Jeh4YBvwtIq7Lqde1ZEHad0jqCZwnqUdE/AxoDtwCjIiIv+TU20/SFiUK0wrMybWBkFT9u+oHrAwQEQOBp4C2EXFDTt1DgasktV7hgdrSzADaAoMkrRMRJwD3A50krQ0g6SDgTOCL0oVpheTk2nCsnP6cC8yvLoyI3wHPSHocQNJPgV8BZ0TE7BUepX1PRHwCnA60Ao6TtG5EnAIEcIqk3wC/Bg6JiHdKGKoVkJNrAyBpLeDCtJXzKdA6LW8BEBFHA+9LmkLyl/iIiHizVPEaSPqBpPOr1yPiU+B8ki6BYyStEhHHAW2Aw4GB/p2VFyfXhqE5MBU4FtgGmJWWN6se2xoRg4D/Aw73X9LSUQpoCWwk6ezqbWkL9gKgK0lLlYg4AtjZv7Py44lbGghJGwP9gROBbsBQYFNgMkk3wWxgn4iYX+NBrOgkVUbEwvTz5iRJ9L2I+ENOnZ4k/xAeEBFfliZSKzYPOs8oSTuQ/H6eiohvIuINSfNJWkQ9Se42jwFWIvlqOdWJtbQk7QwcJel14KWIeFzS5cAvJJ2Xk2B7kfyD+E2pYrXic8s1gyStDDwArAVcASyMiMvSbesABwKdgNsj4qWSBWqLpU9e/Ylk7Gp7YBXgbJL3Zv0A+B1JX/kI4DCSfvE3ShKsrRDuc82giJhFMnb1G2AcsLukWyTtQ9L3eg3J8J4DJDX3nAKllSbW+4FTIuJi4AaSbxRt028do4CDgf+RDLU6xIm1/LnlmiGSVgc+jYiQ1AS4Evh3RDwm6Z9AX5KbWWcBXwGvRMT00kVsaWI9D1gEEBF90/JHSL76jwbeB24DFoT/wjUabrlmhKQ9SG5StU8fGBBJK3XTdAhWX+Aokm6CfYFXnVhLK2d2q2PSpDpB0suSLiGZ8eo+YDrJTa0rSca5WiPhlmsGpJOwnAFcEBEPS2oaEd+kj7COIvlLeUBEDEvrt4yIr0oYcqOXzkh2O/A0cFZEjE3LbwYGAs0j4pu0rBXJPBBTSxWvrXhOriUmqR3JTY99I+I/6Q2rPwK/iYipkgYDP4yIE6uTbkkDNiTtBFwHnAOsBnQEHo6IJ9LtQ0jGsm7vERyNl7sFSiz9ar8X8EdJPySZku7VnFbOa8BOktZzYs2ML4AjI+IfJKM65gO7psPniIgDSfrGHy1ZhFZybrlmRNo18CBwekRctMRg9D8AQyLi3ZIGad8hqSIiFknqQfIIaxPgoYh4Kt3eOSImlTRIKxkn1wyRtAvwF2DLiJglqVlEzCt1XFa3NMEeAqxK8g/hs5Lk0QGNl7sFMiQiRpDMaPWSpHZOrA1HRIwjeYBgCvBuWubE2oi55ZpBkgaQjGXtTfJ31L+kBkJSE9/EMnByzSxJrTyph1nD5eRqZlYE7nM1MysCJ1czsyJwcjUzKwInVzOzInByNTMrAifXRkzSQkmjJb0h6V+SWi7HsXaQVD1r196Sfl9L3VUk/XwZznG2pFPzLV+izi2S9qvHubpL8oTWtsycXBu3ryOiV0RsTPLWg+NyN6YvMq33/yMRMTQiLqqlyipAvZOrWUPi5GrVngbWTVtsb0m6FngF6Cqpn6TnJb2StnBbQTLZjKS3JT1DMoE3afmRkq5OP68m6T5Jr6XLVsBFwDppq/mStN5v0ommX5d0Ts6xzpD0jqRHgfXrughJP0uP85qkfy/RGt9Z0tOS3pW0Z1q/UtIlOec+dnn/IM3AydUASVXAbiRvk4Ukid0WEZsCc4AzgZ0jYjNgJHCKpOYk74raC9gWWL2Gw18FPBkRmwCbAWOB35O8brpXRPwmnXi6B9CH5M2om0vaLn019UEkrxDfF9gij8u5NyK2SM/3FjAoZ1t3YHtgD+Cv6TUMAmZFxBbp8X8maa08zmNWK79au3FrIWl0+vlp4EZgDeCjiHghLe8LbAg8m74HsSnwPLAB8EE6YQmS7gAGL+UcOwJHAKRTKM6S1HaJOv3S5dV0vRVJsm0N3Ff91gVJQ/O4po0lnU/S9dAKGJ6z7e6IWASMk/R+eg39gB/m9MeunJ7b0zvacnFybdy+joheuQVpAp2TWwSMiIiDl6jXCyjUs9MCLoyIvy1xjl8uwzluAfaJiNckHQnskLNtyWNFeu6TIiI3CSOpez3Pa/Yd7hawurwAbC1pXUje3yVpPeBtYK30tTSQvDp6aR4Djk/3rZTUBphN0iqtNhw4Oqcvt7OkjsBTwE8ktZDUmqQLoi6tgSlK3p576BLb9pdUkca8NvBOeu7j0/pIWk/SSnmcx6xWbrlarSLis7QFeKekZmnxmRHxbvp+rwckTQOeATZeyiFOBq6XNAhYCBwfEc9LejYd6vRQ2u/aE3g+bTl/CRwWEa8oeR/VaOAjkq6LuvwBeDGtP4bvJvF3gCdJ3nt1XETMlfR3kr7YV5Sc/DNgn/z+dMxq5lmxzMyKwN0CZmZF4ORqZlYETq5mZkXg5GpmVgROrmZmReDkamZWBE6uZmZF8P+Pz4VkKqaVlAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pass_tf2 = PassiveAggressiveClassifier()\n",
    "pass_tf2.fit(tfidf_train_3, y_train)\n",
    "pred = pass_tf2.predict(tfidf_test_3)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.940\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVcAAAEmCAYAAADWT9N8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3dd5wV1f3G8c+zi4AICAooAgoqKmosiEjssSC2QDQqVkQUNab60xhLokZNTIw1Go3GgiZR1GgkmKjYY1cUCzawEBCkSJGiSPn+/phZvOCWu3Avd/bu8/Y1r71z5syZM6z73bNnzjmjiMDMzAqrotQVMDMrRw6uZmZF4OBqZlYEDq5mZkXg4GpmVgQOrmZmReDgarWStKakf0maI+meVSjnaEmPFLJupSJpN0nvlboelm3yONfyIOko4HRgC2AuMAa4JCKeWcVyjwV+BOwcEYtXuaIZJymA7hExvtR1sYbNLdcyIOl04CrgN8B6wIbAn4D+BSh+I+D9xhBY8yGpSanrYA1ERHhrwBuwNjAPOKyWPM1Igu/kdLsKaJYe2xOYBPwfMA2YAgxOj10IfAUsSq8xBLgA+GtO2V2BAJqk+8cDH5K0nj8Cjs5JfybnvJ2Bl4E56dedc449CVwEPJuW8wjQroZ7q6r/z3PqPwA4AHgfmAmck5O/N/A8MDvNey3QND32dHov89P7PSKn/LOAT4E7qtLSczZJr9Ez3d8AmAHsWer/N7yVdnPLteH7NtAcuL+WPOcCfYDtgG1JAsx5OcfXJwnSnUgC6HWS2kbE+SSt4eER0TIibq6tIpLWAq4B9o+IViQBdEw1+dYBHkzzrgtcATwoad2cbEcBg4EOQFPgjFouvT7Jv0En4FfATcAxwA7AbsCvJG2c5l0C/AxoR/JvtzfwA4CI2D3Ns216v8Nzyl+HpBU/NPfCEfEBSeD9m6QWwK3AbRHxZC31tUbAwbXhWxeYEbX/2X408OuImBYR00lapMfmHF+UHl8UEf8mabVtvpL1WQpsLWnNiJgSEWOryXMgMC4i7oiIxRFxJ/AucHBOnlsj4v2I+AK4m+QXQ00WkfQvLwLuIgmcV0fE3PT6Y4FtACJidES8kF73Y+DPwB553NP5EbEwrc9yIuImYBzwItCR5JeZNXIOrg3fZ0C7OvoCNwAm5OxPSNOWlbFCcF4AtKxvRSJiPsmf0qcAUyQ9KGmLPOpTVadOOfuf1qM+n0XEkvRzVfCbmnP8i6rzJW0maaSkTyV9TtIyb1dL2QDTI+LLOvLcBGwN/DEiFtaR1xoBB9eG73ngS5J+xppMJvmTtsqGadrKmA+0yNlfP/dgRDwcEfuStODeJQk6ddWnqk6frGSd6uN6knp1j4jWwDmA6jin1iE1klqS9GPfDFyQdntYI+fg2sBFxBySfsbrJA2Q1ELSGpL2l/T7NNudwHmS2ktql+b/60pecgywu6QNJa0NnF11QNJ6kr6b9r0uJOleWFJNGf8GNpN0lKQmko4AtgRGrmSd6qMV8DkwL21Vn7rC8anAxt84q3ZXA6Mj4kSSvuQbVrmW1uA5uJaBiLiCZIzrecB0YCLwQ+CfaZaLgVeAN4A3gVfTtJW51ihgeFrWaJYPiBUkow4mkzxB34P0YdEKZXwGHJTm/YzkSf9BETFjZepUT2eQPCybS9KqHr7C8QuAYZJmSzq8rsIk9Qf6kXSFQPJ96Cnp6ILV2BokTyIwMysCt1zNzIrAwdXMrAgcXM3MisDB1cysCMp6EQo1WTPUtFWpq2H1sH2PDUtdBauHCRM+ZsaMGXWNE66XytYbRSz+xkS4asUX0x+OiH6FvH6hlHdwbdqKZpvXOZrGMuTZF68tdRWsHnbZqVfBy4zFX+T9c/vlmOvqml1XMmUdXM2sIRKo4fdYOriaWbYIqKgsdS1WmYOrmWWPCtqNWxIOrmaWMe4WMDMrDrdczcwKTLjlamZWeHLL1cysKDxawMys0PxAy8ys8IS7BczMisItVzOzQnO3gJlZ4Qmo9AMtM7PCc5+rmVmhuVvAzKw43HI1MysCt1zNzApM5TH9teH/ejCz8lNRmd+WB0kfS3pT0hhJr6Rp60gaJWlc+rVtmi5J10gaL+kNST1zyhmU5h8naVCdt7CSt25mViTpA618tvx9JyK2i4iql379AngsIroDj6X7APsD3dNtKHA9JMEYOB/YCegNnF8VkGvi4Gpm2VPVNVDXtvL6A8PSz8OAATnpt0fiBaCNpI7AfsCoiJgZEbOAUUCtb511cDWzbKlazzW/lms7Sa/kbEOrKTGARySNzjm+XkRMAUi/dkjTOwETc86dlKbVlF4jP9Ays4yp1zjXGTl/6tdkl4iYLKkDMErSu7Vf/BuilvQaueVqZtlTwG6BiJicfp0G3E/SZzo1/XOf9Ou0NPskoEvO6Z2BybWk18jB1cyyp0CjBSStJalV1WegL/AWMAKoeuI/CHgg/TwCOC4dNdAHmJN2GzwM9JXUNn2Q1TdNq5G7BcwsW1TQ6a/rAfcraeU2Af4eEQ9Jehm4W9IQ4H/AYWn+fwMHAOOBBcBggIiYKeki4OU0368jYmZtF3ZwNbPsKdAkgoj4ENi2mvTPgL2rSQ/gtBrKugW4Jd9rO7iaWeaoDGZoObiaWaYkb3lxcDUzKywJVTi4mpkVnFuuZmZF4OBqZlYEDq5mZoUmqp9s2sA4uJpZpgi55WpmVgwVFQ1/Zr6Dq5lljluuZmaF5j5XM7PicMvVzKzA/EDLzKxIHFzNzApNeG0BM7NicMvVzKwIHFzNzArMD7TMzIql4cdWB9esePfBC5k7fyFLli5l8ZKl7Hr07zn35AM44ZCdmT5rHgDnXzuCh595mw07rsOY+87j/QnJ24BfevNjfnzJXQA8cO0PWL99a5pUVvLsax/w098OZ+nSWl+vbqto4sSJnDj4OKZO/ZSKigpOGDKUH/74J7zx+uv86LRTmD9vHht17cqtt/+N1q1bA3DZ737LbbfeTGVlJZdfeQ379t2vxHeRIfL0VyuwfkOv5rPZ85dL++Nfn+CqOx77Rt4PJ82gz8BLv5F+zFm3MHf+lwDc+YcTOXTfntzz8OjiVNgAaNKkCZf+/nK279mTuXPnsvNOO7D3Pvty6skncunv/8Buu+/BsFtv4crLL+P8Cy/inbff5p7hd/Hq62OZMnkyB/Tbhzfffp/KyrpfFd1YlEO3QMP/9WDLqQqsTZpUsEaTSpKXWVoxdezYke179gSgVatWbLFFDyZP/oRx77/HrrvtDsBe++zLP+//BwAj//UAhx0xkGbNmtG1Wzc22WRTXn7ppZLVP5OU55ZhDq4ZERH8608/5Nm//ZwTDtllWfopA3fnpeFnc8P5R9Om1ZrL0rt2Wpfn7zyLR/7yE3bZfpPlyhpx3Wn877FLmbdgIfc9+tpquweDCR9/zJgxr7Fj753YcqutGfmvEQDcd+89TJo4EYBPPvmEzp27LDunU6fOTJ78SUnqm1WS8tqyrKjBVdISSWNytq45x66W9Imkipy04yVdm36ukDRM0i1KfCzpzZyyrilm3Ve3vQZfyc5H/Y4BP/wTJx+xG7v03ISb7vkvWx58ATsNvJRPZ3zOpacfAsCnMz5ns/1/xbeP/B1nXX4ft/3meFqt1XxZWd897Tq67XsOzZo2Yc8dNy/VLTU68+bN48jDD+Wyy6+idevW/PmmW/jz9dexc+8dmDdvLk2bNk0yVvPXRNYDxeqUb2DN+r9ZsVuuX0TEdjnbx5AETuB7wERg9xVPUvKvdgOwBnBifP237Xdyyvpxkeu+Wk2ZPgeA6bPmMeLxN9hxq65MmzmXpUuDiOCW+56l19YbAfDVosXMnJP0zb72zkQ+nDSD7ht1WK68hV8tZuRTb3Lwnt9avTfSSC1atIgjDz+UI448mgHfS34Jbr7FFoz8zyM899JoDj/iSLptnPyF0alzZyZNmrjs3E8+mUTHjhuUpN5Z5eC68r4DvAVcDxxZzfGrgXWB4yJi6eqsWCm0aN6Uli2aLfu8z7e3YOwHk1m/XetlefrvtS1vfzAFgHZtW1KRTg/s2mldNt2wPR9NmsFaazZddk5lZQX9dtmS9z6euprvpvGJCE45aQibb9GDn/zs9GXp06YlozmWLl3Kpb+5mJOGngLAgQd9l3uG38XChQv5+KOPGD9+HDv27l2SumeVKpTXlmXFHi2wpqQx6eePIuJ76ecjgTuBB4DfSFojIhalx44C3gH2jIjFK5T3hKQl6edhEXHliheUNBQYCsAaLQt3J0XUYd1WDL/iJACaVFYy/D+vMOq5d7j5ouPYZvPORAQTpszkRxffCcCuPTfll6ceyOIlS1iyJPjRJXcx6/MFdFinFfdedTJN12hCZWUFT738Pjfd+0wpb61ReO7ZZ/n73+5g662/xU47bAfAhRf/hvHjxvHnG64DoP+AQzju+MEAbLnVVhx62OFsv82WNGnShKuuuc4jBVaQ9VZpPlTMp8mS5kVEyxXSmgIfA5tHxFxJ9wE3R8SDko4HjgG2AI6IiGdzzvsY6BURM/K9fkWLDtFs88NX/UZstZn18rWlroLVwy479WL06FcKGgmbrd89Oh+d3yOVD684YHRE9Crk9QulFN0C/YC1gTfTgLkry3cNvAscDgyXtNXqr56ZlZIAKb8ty0oRXI8keUjVNSK6At2AvpJaVGWIiOeAU4AHJW1YgjqaWcmUx2iB1TpDKw2g+wEnV6VFxHxJzwAH5+aNiJGS2gMPSdotTc7tc30jIo5bHfU2s9Ur43EzL0UNriv2t0bEAmCdavIdkrN7W076rcCt6W7XwtfQzDJHLBsN05B5hpaZZYpIgms+W17lSZWSXpM0Mt3vJulFSeMkDU8fsiOpWbo/Pj3eNaeMs9P09yTltcqOg6uZZU6BH2j9hGR4Z5XfAVdGRHdgFjAkTR8CzIqITYEr03xI2hIYCGxF8kD+T5LqHDvn4GpmmVOoB1qSOgMHAn9J9wXsBdybZhkGDEg/90/3SY/vnebvD9wVEQsj4iNgPFDnrA8HVzPLljxbrXm2XK8Cfg5UzfRcF5idM0FpEtAp/dyJZEo+6fE5af5l6dWcUyMHVzPLFCEqKiry2oB2kl7J2YYuK0c6CJgWEaOXK/6boo5jtZ1TIy+WbWaZU4/+1Bm1zNDaBfiupAOA5kBrkpZsG0lN0tZpZ2Bymn8S0AWYJKkJyWSnmTnpVXLPqZFbrmaWOYXoc42IsyOiczpZaSDweEQcDTwBfD/NNohkjROAEek+6fHH0xX5RgAD09EE3YDuQJ2rm7vlambZUvyprWcBd0m6GHgNuDlNvxm4Q9J4khbrQICIGCvpbuBtYDFwWkQs+Waxy3NwNbNMSdYWKGx0jYgngSfTzx9SzdP+iPgSOKyG8y8BLqnPNR1czSxzPP3VzKwIymH6q4OrmWWLymOxbAdXM8uUqvVcGzoHVzPLmOyv1ZoPB1czy5wyiK0OrmaWPW65mpkVmMpksWwHVzPLHLdczcyKoAxiq4OrmWWPW65mZoVW/IVbVgsHVzPLlGSx7IYfXR1czSxzKsqg6ergamaZUwax1cHVzLJF5b5wi6TWtZ0YEZ8XvjpmZlAGXa61tlzH8s03H1btB7BhEetlZo1YWbdcI6JLTcfMzIpFlMcDrbze/ippoKRz0s+dJe1Q3GqZWWNWofy2LKszuEq6FvgOcGyatAC4oZiVMrNGLM/Xame96yCf0QI7R0RPSa8BRMRMSU2LXC8za8QyHjfzkk9wXSSpguQhFpLWBZYWtVZm1mg1pj7X64B/AO0lXQg8A/yuqLUys0ZNym/LsjpbrhFxu6TRwD5p0mER8VZxq2VmjVVjWyy7ElhE0jWQ1wgDM7OV1Si6BSSdC9wJbAB0Bv4u6exiV8zMGi/luWVZPi3XY4AdImIBgKRLgNHAb4tZMTNrvLI+zCof+QTXCSvkawJ8WJzqmFljl4wWKHUtVl1tC7dcSdLHugAYK+nhdL8vyYgBM7PCU/kvll01ImAs8GBO+gvFq46ZWZl3C0TEzauzImZm0Ai6BapI2gS4BNgSaF6VHhGbFbFeZtaIlUPLNZ8xq7cBt5L8QtkfuBu4q4h1MrNGrhBDsSQ1l/SSpNcljU1nmCKpm6QXJY2TNLxqrRRJzdL98enxrjllnZ2mvydpv3zuIZ/g2iIiHgaIiA8i4jySVbLMzApOSiYR5LPVYSGwV0RsC2wH9JPUh2T6/pUR0R2YBQxJ8w8BZkXEpsCVaT4kbQkMBLYC+gF/klRZ18XzCa4LlbTRP5B0iqSDgQ55nGdmtlIqKpTXVptIzEt310i3APYC7k3ThwED0s/9033S43unsa8/cFdELIyIj4DxQO867yGP+/wZ0BL4MbALcBJwQh7nmZmtlHos3NJO0is529Dly1GlpDHANGAU8AEwOyIWp1kmAZ3Sz52AiQDp8TnAurnp1ZxTo3wWbnkx/TiXrxfMNjMrCpHXn/xVZkREr5oORsQSYDtJbYD7gR7VZVt26eqP1ZReq9omEdxfWwERcUhdhZuZ1VsRlhOMiNmSngT6AG0kNUlbp52ByWm2SUAXYJKkJsDawMyc9Cq559SotpbrtfW+g4zZrseGPP3cNaWuhtVD2/5/LHUVrB4Wjp9WlHILMRRLUntgURpY1yRZNvV3wBPA90lGPQ0CHkhPGZHuP58efzwiQtIIkgWrriBZwKo78FJd169tEsFjK31XZmaroEDrmnYEhqVP9iuAuyNipKS3gbskXQy8BlRNmLoZuEPSeJIW60CAiBgr6W7gbWAxcFra3VCrfNdzNTNbLQRUFmCKVkS8AWxfTfqHVPO0PyK+BA6roaxLSCZT5c3B1cwyp1FMf60iqVlELCxmZczMkmFWDT+65vMmgt6S3gTGpfvbSvJTBzMrmgrlt2VZPv3G1wAHAZ8BRMTrePqrmRVRo3j7K1ARERNWaKbX+aTMzGxlCGiS9ciZh3yC60RJvYFIhzT8CHi/uNUys8asDGJrXsH1VJKugQ2BqcCjaZqZWcEpvxWvMi+ftQWmkQ6mNTNbHcogtub1JoKbqGaNgYgYWk12M7NVlvWRAPnIp1vg0ZzPzYHvsfzyW2ZmBZO8Q6vhR9d8ugWG5+5LuoNkXUQzs8ITVBZocYFSWpnpr92AjQpdETOzKqrzDVnZl0+f6yy+7nOtIFkt5hfFrJSZNV6N4tXa6ftjtgU+SZOWRkSdK3Cbma2KcgiutfZspIH0/ohYkm4OrGZWdJLy2rIsn27jlyT1LHpNzMz4ulugoS/cUts7tKreMbMrcJKkD4D5JPceEeGAa2aFp8Isll1qtfW5vgT05Ot3epuZFV1jeKAlgIj4YDXVxcwMKP/pr+0lnV7TwYi4ogj1MbNGT1SU+TjXSqAllMFdmlmDIcq/5TolIn692mpiZgYgaFIGna519rmama1OjaHluvdqq4WZWY6yXhUrImauzoqYmVUpg9i6UqtimZkVjchv6mjWObiaWbaIzK8bkA8HVzPLFAGVDq5mZoXX8EOrg6uZZVAZNFwdXM0sa7K/Vms+HFzNLFPKZbRAOdyDmZWZQryJQFIXSU9IekfSWEk/SdPXkTRK0rj0a9s0XZKukTRe0hu5LwmQNCjNP07SoHzuwcHVzLJFyQytfLY6LAb+LyJ6AH2A0yRtSfKC1cciojvwGF+/cHV/oHu6DQWuhyQYA+cDOwG9gfOrAnJtHFzNLFOqugXy2WoTEVMi4tX081zgHaAT0B8YlmYbxtcvBOgP3B6JF4A2kjoC+wGjImJmRMwCRgH96roP97maWebU44FWO0mv5OzfGBE3VlNeV2B74EVgvYiYAkkAltQhzdYJmJhz2qQ0rab0Wjm4mlnm1GOswIyI6FVrWVJL4B/ATyPi81oCd3UHopb0WrlbwMwyR8pvq7scrUESWP8WEfelyVPTP/dJv05L0ycBXXJO7wxMriW9Vg6uZpYpVdNf89lqLSdpot4MvLPCa6lGAFVP/AcBD+SkH5eOGugDzEm7Dx4G+kpqmz7I6pum1crdAmaWMUKFmQC7C3As8KakMWnaOcClwN2ShgD/Aw5Lj/0bOAAYDywABkOy/Kqki4CX03y/zmdJVgdXM8ucQkzQiohnqLn79hsvA4iIAE6roaxbgFvqc30HVzPLlGQolqe/mpkVVp4Pq7LOwdXMMsfB1Qruyy+/pN8+e7Jw4UIWL17MgO8dyrm/uoC+e+3BvHlzAZg+fRo79NqRu+65nzlz5nDi4GOZNHEiixcv5sc/PZ1jBw0u8V00Du/eMoi5X3zFkqXB4iVL2fWnd9O2ZTPu+EU/NurQmgnTPueYSx9i9ryFAFx+8u7s12sjFixczNArH2XMB9OXldVqzTUYc8MxjHj+Q352w1OluqVM8GLZVhTNmjVj5EOP0rJlSxYtWkTfvXZn3/368cjjX//AHT3w+xx40HcBuPGGP7FFjy25574RTJ8+nR226cERRx5N06ZNS3ULjUq/s+/ns8+/XLZ/xmE78OTrk/jDPaM547AdOOOwHTjv1ufYr9dGbLJBG7Y+6Q56b74e15y2J7uffs+y884/tg//feuTUtxCJhVotEBJeZxrxkiiZcuWACxatIhFixYtNxVw7ty5PP3kExz03QHL8s+bO5eIYP68ebRtuw5Nmvh3Zqkc1Gdj/vroOwD89dF3OLjPxsvS//54kv7Se1NZe61mrN+2BQDbb9qeDm1a8Ohr/ytNpTOoUJMISsnBNYOWLFnCzr17snGX9fnO3vuwY++dlh371wP3s8d39qJ169YAnHzqabz37rt079aZPr225XeXX0lFhb+tq0NE8K+L+vPs1UdwQr+tAOjQpgWfzloAwKezFtC+zZoAbLDuWkyaPm/ZuZ/MmMcG67ZEgkuH7Mo5tzy7+m8gw5Tnf1lWlCaOpHVJlvICWB9YAlR1MG0LvJ5e+yPg2IiYLWlP4IyIOCinnNuAkRFxr6QngY7AF+nh8RHx/WLUv9QqKyt57qVXmT17Nkcdfihvj32LLbfaGoB7776LQYOHLMv72KiH2WbbbXnw4Uf58MMP6H/Afuy8y27Lgq8Vz15n/oMpM+fTfu01GXnxAN6bOKvGvNXNZw+Ckw/chodfmcCkGfOqOatxElCR7biZl6IE14j4DNgOQNIFwLyI+EO6Py8iqo4NIxm0e0meRR8dEa/Una08tGnTht1234NRjzzMllttzWeffcYrr7zM3+++b1meO26/jdPPOAtJbLLJpmzUtRvvv/cuvXbsXcKaNw5TZs4HYPqcLxjx/AfsuPl6TJu9gPXbJq3X9du2YPrspC3wyYx5dG7fctm5ndq1ZMpn89lpi/XZZasNGHrgt1ir+Ro0XaOSeV8u4pe3PVeSe8qG7LdK81Hqvx+fJ4+luxqT6dOnM3v2bAC++OILnnj8MTbbfHMA/nnfPfTb/0CaN2++LH+XLhvy1BOPAzBt6lTGjXuPrt02Xv0Vb2RaNGtCyzXXWPZ5n54bMnbCZzz44kccs08PAI7ZpwcjX/gQgAdf/Iij9krSe2++Hp/P/4pPZy1g8B8eYbPBt7HFCcM4+5Zn+Ptj7zbywEq6WHZ+W5aV7MmHpEqSKWg31+O0v0mq6hYYFRFnFr5mpTX10ymcfOJglixZwtKlSznk0MPY/4Ckp+Teu+/m9DN/vlz+s84+j1NOGsxOO2xLRPDri39Lu3btSlH1RqVD2xYMP/dAAJpUiuFPvc+o0f9j9PvT+Osv+jFo3y2ZOH0uR//2PwA89PLH7NdrI8b+5TgWLFzEyVc+VlvxjVrSLZDxyJkHJdNpi3iBb3YLLAHeBLoCo4G+EbFE0h7AmdX0uf4rIv6R9rmeUVe3gKShJK9ooEuXDXd4e9xHBb8nK572h1xX6ipYPSx85jKWzvlfQSNhj29tH7fe/0Reeb/dve3outZzLZVSdAt8kfa5bgQ05euFEj4DVnwvzTrAjPoUHhE3RkSviOjVrn37Va6smZWA8twyrGR9rhExB/gxcEa6oO04YANJPQAkbUQysmBMzaWYWTnyUKxVFBGvSXodGBgRd0g6BrhVUnNgEXBiGoSr5Pa5zoiIfVZ3nc2s+LL+sCofRQ+uEXHBCvstV9g/OOfzsySvwK2unD2LUD0zyyIHVzOzwkq6Uxt+dHVwNbNsaQDrBuTDwdXMMqcMYquDq5llUBlEVwdXM8sYlcUMLQdXM8uUBjA/IC8OrmaWPWUQXR1czSxzPBTLzKwIyqDL1cHVzLKnDGKrg6uZZYyqfy1OQ+PgamaZItwtYGZWFGUQWx1czSyDyiC6OriaWeZ4KJaZWRGUw2LZpX61tpnZNxXoHVqSbpE0TdJbOWnrSBolaVz6tW2aLknXSBov6Q1JPXPOGZTmHydpUD634OBqZplStVh2gd6hdRvQb4W0XwCPRUR34LF0H2B/oHu6DQWuhyQYA+cDOwG9gfOrAnJtHFzNLFvSxbLz2eoSEU8DM1dI7g8MSz8PAwbkpN8eiReANpI6AvsBoyJiZkTMAkbxzYD9De5zNbPMKXKX63oRMQUgIqZI6pCmdwIm5uSblKbVlF4rB1czy578o2s7Sa/k7N8YETcW8KpRS3qtHFzNLGPqtVj2jIjoVc8LTJXUMW21dgSmpemTgC45+ToDk9P0PVdIf7Kui7jP1cwyJd+BAqvQdTACqHriPwh4ICf9uHTUQB9gTtp98DDQV1Lb9EFW3zStVm65mln2FKjTVdKdJK3OdpImkTz1vxS4W9IQ4H/AYWn2fwMHAOOBBcBggIiYKeki4OU0368jYsWHZN/g4GpmmVOoGVoRcWQNh/auJm8Ap9VQzi3ALfW5toOrmWWOV8UyMyuCMoitDq5mljFeLNvMrPC8WLaZWZGUQWx1cDWz7HHL1cysCLxYtplZEbjlamZWYPkuJ5h1Dq5mljnuFjAzK4aGH1sdXM0se8ogtjq4mln2uM/VzKzAVL/FsjPLi2WbmRWBW65mljll0HB1cDWz7PFQLDOzQvMkAjOzwlvFlw9mhoOrmWWOF8s2MyuCMoitDq5mlj1lEFsdXM0sg8ogujq4mlnmlMNQLEVEqetQNJKmAxNKXY8iaAfMKHUlrF7K9Xu2UUS0L2SBkh4i+ffKx72IOyEAAAhBSURBVIyI6FfI6xdKWQfXciXplYjoVep6WP78PWt8vLaAmVkROLiamRWBg2vDdGOpK2D15u9ZI+M+VzOzInDL1cysCBxczcyKwMG1DEhat9R1sPqRtKuk7UpdDyseB9cGTlJf4CpJbVUOSwk1ApL2A64Fmpa6LlY8Dq4NWBpYLwNujohZeDpz5qWB9VLgJxHxkqT2kjYpdb2s8BxcGyhJ/UgC68kR8aSkLsA5kvKdNmirmaRNSYZkXRERT0nqDNwNbF7amlkxOLg2XDsBLSLiBUntgfuBaRFRjvPXGzxJa0bEeOARYDtJuwHDgfsi4t+lrZ0Vg4NrAyNpF0nnRMSFwHOSngdGAn+OiOtz8nUpWSVtOZJ6ABdJ6h4RJwHNgduAURHxx5x835e0Y4mqaQXm4NpASKr6XvUF1gaIiEHA00DbiLgpJ+/RwDWSWq32ilp1ZgFtgSGSNomI04AHgI6SNgaQNBA4D/i8dNW0QnJwbTjWTr9+CSyqSoyIs4BnJD0BIOlQ4GfAuRExd7XX0r4hIj4FzgFaAqdI2jQiTgcCOF3SmcD/AUdFxHslrKoVkINrAyCpG/DbtJUzFWiVpq8JEBEnAB9KmkLyQ3xcRLxdqvoaSPqWpIur9iNiKnAxSZfAiZLaRMQpQGvgWGCQv2flxcG1YWgOTANOBnYF5qTpzarGtkbEEOBy4Fj/kJaOUkALYCtJF1QdS1uwlwBdSFqqRMRxwD7+npUfL9zSQEjaGugH/BDYEBgBbA9MJukmmAsMiIhFNRZiRSepMiKWpJ93IAmiH0TEL3Py9CD5RXh4RMwrTU2t2DzoPKMk7Uny/Xk6Ir6KiLckLSJpEfUgedr8JrAWyZ+W0xxYS0vSPsBgSW8AL0XEE5KuBH4s6aKcALsdyS/Er0pVVys+t1wzSNLawINAN+AqYElEXJEe2wQ4AugI3BERL5WsorZMOvPqNyRjV9sBbYALSN6b9S3gLJK+8lHAMST94m+VpLK2WrjPNYMiYg7J2NWvgHHAAZJukzSApO/1OpLhPYdLau41BUorDawPAKdHxO+Bm0j+omib/tUxGjgSeJxkqNVRDqzlzy3XDJG0PjA1IkLSGsDVwD8i4jFJfwf6kDzMOh9YALwaETNLV2NLA+tFwFKAiOiTpj9C8qf/GOBD4HZgcfgHrtFwyzUjJB1I8pCqXTphQCSt1O3TIVh9gMEk3QSHAK85sJZWzupWJ6ZBdaKklyVdRrLi1f3ATJKHWleTjHO1RsIt1wxIF2E5F7gkIh6S1DQivkqnsI4m+aE8PCJGpvlbRMSCEla50UtXJLsD+C9wfkSMTdNvBQYBzSPiqzStJck6ENNKVV9b/RxcS0zSOiQPPQ6JiH+mD6x+BZwZEdMkDQW2iYgfVgXdklbYkLQ3cD1wIbAe0AF4KCKeTI8PJxnLuodHcDRe7hYosfRP+4OBX0nahmRJutdyWjmvA3tL2syBNTM+B46PiL+RjOpYBOyXDp8jIo4g6Rt/tGQ1tJJzyzUj0q6BfwPnRMSlKwxG/yUwPCLeL2klbTmSKiJiqaTuJFNY1wD+ExFPp8c7RcQnJa2klYyDa4ZI2hf4I7BTRMyR1CwiFpa6Xla3NMAeBaxL8ovwWUny6IDGy90CGRIRo0hWtHpJ0joOrA1HRIwjmUAwBXg/TXNgbcTccs0gSf1JxrL2IvkZ9TepgZC0hh9iGTi4Zpakll7Uw6zhcnA1MysC97mamRWBg6uZWRE4uJqZFYGDq5lZETi4mpkVgYNrIyZpiaQxkt6SdI+kFqtQ1p6Sqlbt+q6kX9SSt42kH6zENS6QdEa+6SvkuU3S9+txra6SvKC1rTQH18bti4jYLiK2JnnrwSm5B9MXmdb7/5GIGBERl9aSpQ1Q7+Bq1pA4uFqV/wKbpi22dyT9CXgV6CKpr6TnJb2atnBbQrLYjKR3JT1DsoA3afrxkq5NP68n6X5Jr6fbzsClwCZpq/myNN+Z6ULTb0i6MKescyW9J+lRYPO6bkLSSWk5r0v6xwqt8X0k/VfS+5IOSvNXSros59onr+o/pBk4uBogqQmwP8nbZCEJYrdHxPbAfOA8YJ+I6Am8ApwuqTnJu6IOBnYD1q+h+GuApyJiW6AnMBb4BcnrpreLiDPThae7A71J3oy6g6Td01dTDyR5hfghwI553M59EbFjer13gCE5x7oCewAHAjek9zAEmBMRO6blnySpWx7XMauVX63duK0paUz6+b/AzcAGwISIeCFN7wNsCTybvgexKfA8sAXwUbpgCZL+Cgyt5hp7AccBpEsozpHUdoU8fdPttXS/JUmwbQXcX/XWBUkj8rinrSVdTNL10BJ4OOfY3RGxFBgn6cP0HvoC2+T0x66dXtvLO9oqcXBt3L6IiO1yE9IAOj83CRgVEUeukG87oFBzpwX8NiL+vMI1froS17gNGBARr0s6Htgz59iKZUV67R9FRG4QRlLXel7XbDnuFrC6vADsImlTSN7fJWkz4F2gW/paGkheHV2dx4BT03MrJbUG5pK0Sqs8DJyQ05fbSVIH4Gnge5LWlNSKpAuiLq2AKUrennv0CscOk1SR1nlj4L302qem+ZG0maS18riOWa3ccrVaRcT0tAV4p6RmafJ5EfF++n6vByXNAJ4Btq6miJ8AN0oaAiwBTo2I5yU9mw51+k/a79oDeD5tOc8DjomIV5W8j2oMMIGk66IuvwReTPO/yfJB/D3gKZL3Xp0SEV9K+gtJX+yrSi4+HRiQ37+OWc28KpaZWRG4W8DMrAgcXM3MisDB1cysCBxczcyKwMHVzKwIHFzNzIrAwdXMrAj+HzfJT3Ab8eBMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "log_tf2 = LogisticRegression()\n",
    "log_tf2.fit(tfidf_train_3, y_train)\n",
    "pred = log_tf2.predict(tfidf_test_3)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "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>00</th>\n",
       "      <th>02</th>\n",
       "      <th>025</th>\n",
       "      <th>0330</th>\n",
       "      <th>0930</th>\n",
       "      <th>10</th>\n",
       "      <th>100</th>\n",
       "      <th>1000</th>\n",
       "      <th>10000</th>\n",
       "      <th>100000</th>\n",
       "      <th>...</th>\n",
       "      <th>zones</th>\n",
       "      <th>zoo</th>\n",
       "      <th>zschaepe</th>\n",
       "      <th>zuckerberg</th>\n",
       "      <th>zulia</th>\n",
       "      <th>zuma</th>\n",
       "      <th>zumas</th>\n",
       "      <th>zurich</th>\n",
       "      <th>état</th>\n",
       "      <th>žižek</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 20974 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   00  02  025  0330  0930  10  100  1000  10000  100000  ...  zones  zoo  \\\n",
       "0   0   0    0     0     0   0    0     0      0       0  ...      0    0   \n",
       "1   0   0    0     0     0   0    0     0      0       0  ...      0    0   \n",
       "2   0   0    0     0     0   0    0     0      0       0  ...      0    0   \n",
       "3   0   0    0     0     0   0    0     0      0       0  ...      0    0   \n",
       "4   0   0    0     0     0   0    0     0      0       0  ...      0    0   \n",
       "\n",
       "   zschaepe  zuckerberg  zulia  zuma  zumas  zurich  état  žižek  \n",
       "0         0           0      0     0      0       0     0      0  \n",
       "1         0           0      0     0      0       0     0      0  \n",
       "2         0           0      0     0      0       0     0      0  \n",
       "3         0           0      0     0      0       0     0      0  \n",
       "4         0           0      0     0      0       0     0      0  \n",
       "\n",
       "[5 rows x 20974 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "count_df = pd.DataFrame(count_train.A, columns=count_vectorizer.get_feature_names())\n",
    "count_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FAKE -15.726725920823498 video\n",
      "FAKE -8.203009566802208 breaking\n",
      "FAKE -7.925680913325372 hillary\n",
      "FAKE -7.886026956177907 gop\n",
      "FAKE -7.453474202930462 watch\n",
      "FAKE -5.257061404941885 racist\n",
      "FAKE -4.659838779633869 tweets\n",
      "FAKE -4.439127182902908 america\n",
      "FAKE -4.21994326414817 won\n",
      "FAKE -3.9224814595067192 details\n",
      "\n",
      "TRUE 6.497287683258772 trumps\n",
      "TRUE 5.528881975282868 says\n",
      "TRUE 4.961962816871995 factbox\n",
      "TRUE 3.5093921743672136 exclusive\n",
      "TRUE 3.4058444729756503 urges\n",
      "TRUE 3.290121220768436 islamic state\n",
      "TRUE 3.253558902201538 rohingya\n",
      "TRUE 3.057140858761049 seek\n",
      "TRUE 3.0429010749600702 lawmakers\n",
      "TRUE 2.9478581964430064 german\n"
     ]
    }
   ],
   "source": [
    "def most_informative_feature_for_binary_classification(vectorizer, classifier, n=100):\n",
    "    \"\"\"\n",
    "    See: https://stackoverflow.com/a/26980472\n",
    "    \n",
    "    Identify most important features if given a vectorizer and binary classifier. Set n to the number\n",
    "    of weighted features you would like to show. (Note: current implementation merely prints and does not \n",
    "    return top classes.)\n",
    "    \"\"\"\n",
    "\n",
    "    class_labels = classifier.classes_\n",
    "    feature_names = vectorizer.get_feature_names()\n",
    "    topn_class1 = sorted(zip(classifier.coef_[0], feature_names))[:n]\n",
    "    topn_class2 = sorted(zip(classifier.coef_[0], feature_names))[-n:]\n",
    "\n",
    "    for coef, feat in topn_class1:\n",
    "        print(class_labels[0], coef, feat)\n",
    "\n",
    "    print()\n",
    "\n",
    "    for coef, feat in reversed(topn_class2):\n",
    "        print(class_labels[1], coef, feat)\n",
    "\n",
    "\n",
    "most_informative_feature_for_binary_classification(tfidf_vectorizer, pass_tf, n=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FAKE -15.726725920823498 landscape\n",
      "FAKE -8.203009566802208 arpaio arresting\n",
      "FAKE -7.925680913325372 court justice nomination\n",
      "FAKE -7.886026956177907 conference wednesday white\n",
      "FAKE -7.453474202930462 leads criminal\n",
      "FAKE -5.257061404941885 genealogy trump uturn\n",
      "FAKE -4.659838779633869 killed 200\n",
      "FAKE -4.439127182902908 additional barriers\n",
      "FAKE -4.21994326414817 libs twitter nuts\n",
      "FAKE -3.9224814595067192 called khizr\n",
      "\n",
      "TRUE 6.497287683258772 kellyanne conway presidents\n",
      "TRUE 5.528881975282868 harry recent engagement\n",
      "TRUE 4.961962816871995 chrischristie busted beach\n",
      "TRUE 3.5093921743672136 chief declines comment\n",
      "TRUE 3.4058444729756503 korea security jitters\n",
      "TRUE 3.290121220768436 debate whiniest\n",
      "TRUE 3.253558902201538 guess spotted vip\n",
      "TRUE 3.057140858761049 hezbollah disarms\n",
      "TRUE 3.0429010749600702 destroyed life\n",
      "TRUE 2.9478581964430064 commitments bans\n"
     ]
    }
   ],
   "source": [
    "most_informative_feature_for_binary_classification(tfidf_vectorizer_3, pass_tf, n=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hashing Vectorizer (Exploring parameters n_gram)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "hash_vectorizer = HashingVectorizer(stop_words='english',ngram_range=(1,2))\n",
    "hash_train = hash_vectorizer.fit_transform(X_train)\n",
    "hash_test = hash_vectorizer.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.956\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clf = PassiveAggressiveClassifier()\n",
    "clf.fit(hash_train, y_train)\n",
    "pred = clf.predict(hash_test)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:   0.938\n",
      "Confusion matrix, without normalization\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clf = LogisticRegression()\n",
    "clf.fit(hash_train, y_train)\n",
    "pred = clf.predict(hash_test)\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"accuracy:   %0.3f\" % score)\n",
    "cm = metrics.confusion_matrix(y_test, pred, labels=['FAKE', 'TRUE'])\n",
    "plot_confusion_matrix(cm, classes=['FAKE', 'TRUE'])"
   ]
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}