Analisis de datos.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "plt.rcParams['figure.figsize'] = [16, 6]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_trolls = pd.read_csv(\"dataset_troll_all.csv\")\n",
    "df_users = pd.read_csv(\"dataset_users_trial.csv\")\n",
    "df = pd.concat([df_trolls, df_users])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((166688, 6), (24760, 6))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_trolls.shape, df_users.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>content</th>\n",
       "      <th>followers</th>\n",
       "      <th>following</th>\n",
       "      <th>retweet</th>\n",
       "      <th>created_at</th>\n",
       "      <th>troll</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>70349</th>\n",
       "      <td>RT @pabelml: INAudito: - La compañera que reco...</td>\n",
       "      <td>7.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-04-18 15:48:01</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10898</th>\n",
       "      <td>@SCC1705 https://t.co/js9QoU770J</td>\n",
       "      <td>39.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>False</td>\n",
       "      <td>2019-07-18 16:41:48</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52583</th>\n",
       "      <td>@LaCarreno189 @miligonzzz MILI dijo el COMICO ...</td>\n",
       "      <td>62.0</td>\n",
       "      <td>184.0</td>\n",
       "      <td>False</td>\n",
       "      <td>2019-08-12 01:28:15</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71315</th>\n",
       "      <td>@Elvetadoportui1 @MashiRafael Calla mmv</td>\n",
       "      <td>21.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>False</td>\n",
       "      <td>2019-06-15 17:05:09</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3253</th>\n",
       "      <td>@JorgeGlasLibre @MashiRafael @JorgeGlas @ecuai...</td>\n",
       "      <td>8.0</td>\n",
       "      <td>87.0</td>\n",
       "      <td>False</td>\n",
       "      <td>2019-04-22 17:17:15</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 content  followers  \\\n",
       "70349  RT @pabelml: INAudito: - La compañera que reco...        7.0   \n",
       "10898                   @SCC1705 https://t.co/js9QoU770J       39.0   \n",
       "52583  @LaCarreno189 @miligonzzz MILI dijo el COMICO ...       62.0   \n",
       "71315            @Elvetadoportui1 @MashiRafael Calla mmv       21.0   \n",
       "3253   @JorgeGlasLibre @MashiRafael @JorgeGlas @ecuai...        8.0   \n",
       "\n",
       "       following retweet           created_at troll  \n",
       "70349       22.0    True  2019-04-18 15:48:01  True  \n",
       "10898       89.0   False  2019-07-18 16:41:48  True  \n",
       "52583      184.0   False  2019-08-12 01:28:15  True  \n",
       "71315       88.0   False  2019-06-15 17:05:09  True  \n",
       "3253        87.0   False  2019-04-22 17:17:15  True  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sample(5).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "users_copy = df_users.copy()\n",
    "trolls_copy = df_trolls.copy()\n",
    "n = 10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "users_copy['created_at'] = users_copy['created_at'].apply(pd.to_datetime)\n",
    "trolls_copy['created_at'] = trolls_copy['created_at'].apply(pd.to_datetime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "troll_range = trolls_copy[\n",
    "    (trolls_copy.created_at >= '2019-01-01') &\n",
    "    (trolls_copy.created_at <= '2019-08-10')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "users_range = users_copy[\n",
    "    (users_copy.created_at >= '2019-01-01') &\n",
    "    (users_copy.created_at <= '2019-08-10')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10780, 6)"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "users_range.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(10,7))\n",
    "ax.set_title('Histogram of tweet post dates (samples of 10000 tweets)')\n",
    "ax.set_ylabel('Number of tweets')\n",
    "ax.set_xlabel('Post date')\n",
    "\n",
    "nontroll_dates = users_range.created_at.sample(n)\n",
    "troll_dates = troll_range.created_at.sample(n)\n",
    "\n",
    "ax.hist(nontroll_dates, bins=10, label='non-troll')\n",
    "ax.hist(troll_dates, bins=10, label='troll', alpha=0.5)\n",
    "\n",
    "ax.legend()\n",
    "ax.tick_params(axis='x', rotation=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Followers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(10,7))\n",
    "fig.tight_layout()\n",
    "\n",
    "ax.set_title('Histogram of followers (samples of 100K tweets)')\n",
    "ax.set_ylabel('Log Density')\n",
    "ax.set_xlabel('Number of followers (x $10^7$)')\n",
    "\n",
    "troll_followers = df_trolls.followers.sample(n)\n",
    "nontroll_followers = df_users.followers.sample(n)\n",
    "\n",
    "ax.hist(troll_followers, bins=100, label='troll', density=True, log=True)\n",
    "ax.hist(nontroll_followers, bins=100, label='non-troll', density=True, log=True, alpha=0.2)\n",
    "\n",
    "ax.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Following"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(10,7))\n",
    "ax.set_title('Histogram of following (samples of 1000 tweets)')\n",
    "ax.set_ylabel('Density')\n",
    "ax.set_xlabel('Number of following')\n",
    "df_trolls.following.sample(n).hist(ax=ax, bins=100, label='troll', color='blue', alpha=1, density=True)\n",
    "df_users.following.sample(n).hist(ax=ax, bins=100, label='non-troll', color='orange', alpha=0.5, density=True)\n",
    "ax.set_xlim(0, 1000)\n",
    "ax.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Posts dates distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['created_at'] = pd.to_datetime(pd.to_datetime(df['created_at']))\n",
    "df_trolls['created_at'] = pd.to_datetime(pd.to_datetime(df_trolls['created_at']))\n",
    "df_users['created_at'] = pd.to_datetime(pd.to_datetime(df_users['created_at']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### All users interactions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x360 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(ncols=7, figsize=(30,5))\n",
    "plt.subplots_adjust(wspace=0.05)  #Remove some whitespace between subplots\n",
    "\n",
    "for idx, gp in df.groupby(df['created_at'].dt.dayofweek):\n",
    "    ax[int(idx)].set_title(gp['created_at'].dt.day_name().iloc[0])  #Set title to the weekday\n",
    "\n",
    "    (gp.groupby(gp['created_at'].dt.hour).size().rename_axis('Tweet Hour').to_frame('')\n",
    "        .reindex(np.arange(0,24,1)).fillna(0)\n",
    "        .plot(kind='bar', ax=ax[int(idx)], rot=0, ec='k', legend=False))\n",
    "\n",
    "    # Ticks and labels on leftmost only\n",
    "    if idx == 0:\n",
    "        _ = ax[int(idx)].set_ylabel('Counts', fontsize=11)\n",
    "\n",
    "    _ = ax[int(idx)].tick_params(axis='both', which='major', labelsize=7,\n",
    "                            labelleft=(idx == 0), left=(idx == 0))\n",
    "\n",
    "# Consistent bounds between subplots. \n",
    "lb, ub = list(zip(*[axis.get_ylim() for axis in ax]))\n",
    "for axis in ax:\n",
    "    axis.set_ylim(min(lb), max(ub)) \n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Trolls interactions by day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x360 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(ncols=7, figsize=(30,5))\n",
    "plt.subplots_adjust(wspace=0.05)  #Remove some whitespace between subplots\n",
    "\n",
    "for idx, gp in df.groupby(df_trolls['created_at'].dt.dayofweek):\n",
    "    ax[int(idx)].set_title(gp['created_at'].dt.day_name().iloc[0])  #Set title to the weekday\n",
    "\n",
    "    (gp.groupby(gp['created_at'].dt.hour).size().rename_axis('Tweet Hour').to_frame('')\n",
    "        .reindex(np.arange(0,24,1)).fillna(0)\n",
    "        .plot(kind='bar', ax=ax[int(idx)], rot=0, ec='k', legend=False))\n",
    "\n",
    "    # Ticks and labels on leftmost only\n",
    "    if idx == 0:\n",
    "        _ = ax[int(idx)].set_ylabel('Counts', fontsize=11)\n",
    "\n",
    "    _ = ax[int(idx)].tick_params(axis='both', which='major', labelsize=7,\n",
    "                            labelleft=(idx == 0), left=(idx == 0))\n",
    "\n",
    "# Consistent bounds between subplots. \n",
    "lb, ub = list(zip(*[axis.get_ylim() for axis in ax]))\n",
    "for axis in ax:\n",
    "    axis.set_ylim(min(lb), max(ub)) \n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### No trolls interactions by day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABroAAAFJCAYAAADEyv2xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzdeZhsV10v/O8vBMIckATSptNhVkBEMDgBL6D48gIBGURm3iCXCBe4vjIFBRXEgYh6ZRAwoswRSAQu5BpnQgBRCXl5EEFFhJNzTnIwIGNAQpJ1/9j7JN2VHqpPurpqV30+z7Of7tq1eu1V1fWrvfb+7bV2tdYCAAAAAAAAQ3PYtBsAAAAAAAAAh0KiCwAAAAAAgEGS6AIAAAAAAGCQJLoAAAAAAAAYJIkuAAAAAAAABkmiCwAAAAAAgEGS6OIaq6qTquqD875NYGNV9aKqesu02wFcM1W1r6ruM+12wKLQjwZGVdV9qmrftNsB884+GIZNDDNKomtBVNXPV9Wfjqz79AbrHr27rQN2WlV9rqouraqjRtZ/rKpaVd1yOi0D1lNVX1+1XFFV31z1+HHTbh8sMv1omH9Vdc+q+tuq+kpV/WdVfaiq7j7G332uqu63G22ERWQfDMMmhtlNEl2L49wk96iqayVJVR2T5NpJ7jay7rZ9WWD4PpvkMQcfVNWdk1xves0BNtJau+HBJckFSR68at1bp90+WHD60TDHqurGSc5K8sok35Hk2CQvTvKtXdj24ZPeBgycfTAMmxhm10h0LY6PpPsi+b7+8f+V5H1J/mVk3WdaaxdW1XdX1V/2V7P9S1X91MGKqupmVfWeqvpqVf1Dktus3lA/WuSpfTb+S1X1e1VVq57/6ar6VP/cn1fV8f36qqr/WVX/0V9J9/Gq+p4xt/nyqtrbP//RqrpXv/6YqvpGVd1sVdnvr6qLq+raO/HGwgx7c5Inrnr8/yZ508EHVXVkVb2pj4c9VfXCqjqsf+6kqvpgVf1WH6ufraoHrPrbW1XV+6vqa1X1l0lGR46dUVUH+lg+t6ru1K+/e1V9fvVBfVU9oqo+Npm3AOZDVb2lql606vH9qupzqx4vV9W7+nj+bFU9fdVzP1RV5/f7yM9X1ctWPXdSH/9fqKrnj2zzh6vq76rqy1V1UVW94uC+s6p+v6pOHSl/dlU9Y+dfPUydfvRVZfWjmUe3T5LW2h+31i5vrX2ztfYXrbWPV9VtqupvquqL/b7yrVV1kySpqjcnWUny3upGYD+v1pl2sFaN+qpuuu8z+/36V5OcVFXXq6o39HH9ySR3H/n751fVZ/p+9yer6mH9+iP675k7ryp78+pGhR89yTcMdpF98FVl7YMZIjF8VVkxPGESXQuitXZpkr9P9+WR/ucHknxwZN25VXWDJH+Z5PQkN083IuTV1Z+oTvJ7Sf4ryVKSn+6XUSem66DfJclPJbl/klTVQ5P8QpKHJzm6b8Mf93/zf/dtuH2SmyR5VJIvjrnNj6T7gvyOvt1nVNV1W2sHkpzTt+Ggxyd5W2vt2xu8XTAv/i7JjavqDtVdKfOoJKvvo/XKJEcmuXWSe6dLij1p1fM/mK7zcVSS30zyh6s6Cacn+Wj/3EvSJdFWOzvJ7dJ9h5yf5K1J0lr7SLq4/vFVZR+fLikHHII+vs9Kty88Nl18Pbeqfqwv8sokL2ut3TjdlXJn9n935ySvSvLY/u++M8kxq6q+LMnPpovzeyT5f5L8TP/cG5M8tq5Kjt8i3ffI2ybzKmF69KP1o5l7/5rk8qp6Y1U9oKpuuuq5SvIb6faRd0hyXJIXJUlr7QlZOwr7N8fc3k+k2xffJF0f+ZfTnTi7Tbp4H+1XfybJvdL121+c5C1VtdRa+1a6/e7jV5V9TJK/aq1dPGZbYKbZB9sHM2xiWAzvJomuxfL+XPUlcq90Qf2BkXXvT/el8LnW2utba5e11s5P8idJfrI/mfaIJL/UWruktfaJdCe7Rr20tfbl1toF6TL1B7P0P5PkN1prn2qtXZbk15N8X59F/3aSGyX57iTVl7lonG221t7SWvti397fTnJEku/qn35j+s5/X9dj4qQ6i+PgqK4fT/LPSfb36w8mvn6+tfa11trnkvx2kies+ts9rbU/aK1dni6OlpLcoqpW0nUcfrG19q3W2rlJ3rt6o621P+rr/Va6kwF3qaoj+6dXx+R3pOt4nL6zLxsWyg8luXFr7ddba5e21v4tyR8mOTjH+beT3K6qbtbH5d/36x+Z5N2ttQ/1sfoL6U7oJekS0621v+/3rf+e5LR0yay01v42yTcPPs5VJ9a+MOHXCtOiH60fzZxqrX01yT2TtCR/kOTi/urtW7TW/q219pd9n/fiJL+Tq/Z9h+rDrbV3t9auaK19M91JsF9rrf1na21vkleMtO+M1tqFffm3J/l0kh/on15z4Um6vrwYZd7YB9sHM2xiWAzvComuxXJuknv2V6gd3Vr7dJK/TfIj/brv6cscn+QHq5uq6MtV9eUkj0t3lffRSQ5PsndVvXvW2daBVb9/I8kN+9+PT/LyVfX+Z7qTase21v4m3ZXlv5fk81V1WnXzpW+5zap6dj/89Ct9vUfmqqnU/leSO1bVrdOd7P9Ka+0fxnrHYPjenG60xklZNW1huvi4TtbG0p50ozoOujKOW2vf6H+9YborWr/UWrtk5G+TdDvwqnppdVOsfDXJ51ZtM+lGlT24qm6Y7sD+A621iw7p1QFJt29dGdlvPy9Xjc56UpI7JvmXqvqHqnpgv/47s2rf2lr7err9cpKknzbif1c3DelXk/xK1k5T+qZcdRW5kZnMO/1o/WjmWH9S66TW2nK6eP7OJL9b3VSAb6uq/f2+8C0ZmbL7EOwdefyd2TxGn1hVH1sV+99zsA39xSuXJLl3VX13upHb77mG7YNZYx9sH8ywiWExvCskuhbLh9MF3MlJPpRcefXahf26C1trn00XwO9vrd1k1XLD1trTklycbiqj41bVu7KNNuxN8jMjdV+vvzI8rbVXtNa+P8md0g0Zfe5W26xu/tNT0p0wv2lr7SZJvpL+qvTW2n8leUe6L0dXuLFQWmt7knw2yQOTvHPVU19Id9XK8avWreSqEV+buSjJTfth5av/9qDHppuS5X7pvnNu2a8/GJP7030fPSxiEsZ1SZLrr3q8eorBvUk+PbJvvVFr7cFJ0lr7l9bao9NN//DbSf6kqq6bLpav3Lf2yefvWFXv7yf5RJLbtm7aw1/KqhFf6WL34VV113TTLa0Z2QlzRj/aPpsF0Vr75yRvSHfi7TfSjfT63n5f+Pis3Re2kT9fs7/ur+AevV/W6N+s2R9nbYwen26U2TOS3KyP0U+MtOHgFeNPSHJmH7cwT+yD7YMZNjEshneFRNcCad20COcleVa6IaIHfbBfd27/+Kwkt6+qJ1TVtfvl7lV1h9ZNYfbOJC+qqutX1R1z9TnEN/PaJD9/cH7Vqjqyqh7Z/373qvrB6m7Kd0m6OVAvH2ObN0r3xXNxksOr6peS3Hhku29KN6LlIVl7jyJYBE9O8qMjI7AuT7fD/bWqulF/EP2sjBEfffLsvCQvrqrrVNU9kzx4VZEbJflWujmNr59uSPioN6UbcXLnJO/a/kuChfOxJA+qqptW1VKS/7HquQ8nubS/muy6/ajKO1fV9ydJvz8/qrV2RbqOd0tyRZIzkvxEVf1wVR2R5Fez9uTbjfryl1TVHXLV/bmSXPld8LF0J9jOcGKNeaYfrR/N/OpHMD+7qpb7x8elm17o79LFyNeTfLmqjk134mu1z6e73+1B/5rkulX1oD4eX5huGqPNvCNdbN+0b8MzVz13g3T75ov7tj0pXQJutTenu4Ds8Vk7gwPMBftg+2CGTQyL4d0i0bV43p/uiu4Prlr3gX7duUnSWvtauhvxPTpddv1AklNzVQf9GemGfh5Id6Xb68fdeGvtXX1db6tu6odPJHlA//SN012t9qV0Q0G/mOS3xtjmnyc5O91BxZ50X0hrpoNorX0o3Um981t3LyJYGK21z7TWzlvnqWem24n/e7rvhNOT/NGY1T42yQ+mG+79y1l7UP2mdLG4P8kn050kGPWudKPJ3jWSgAPW94Ykn0oXW3+W7ubzSZLWzTH+wHT36/hcuhGbv5+rOtkPTPKpqvpauv3qo1p3L6+PJ/nZdCfY9qfbx66e6uHZ6TryX+vre/s67XpjuoS1q9NYBPrR+tHMp6+l69f+fVVdkq7v+ol0+8EXJ7lbugs//nfWzpCQdCO+XthPhfSc1tpXkvz3JK9Lt2+9JMm+Lbb/4nTx99kkf5FV+9TW2ifTjcb+cLqk2p3TXw2/qsy+JOenS4itPoEI88Q+2D6YYRPDYnjiqrXRUfMwn6rqb5Kc3lp73bTbAiRV9Zl0Q8f/atptAQ5NVf1okj9McuumUwlzSz8aZltV/VG6qZ9eOO22ADvLPhiGTQzvnsOn3QDYDVV193RX4v3EtNsCJFX1iHRXnf7NtNsCHJqquk66EWF/IMkF80s/GmZbVd0yycOT3HW6LQF2mn0wDJsY3l2mLmTuVdUbk/xVkv+vHwYLTFFVnZPkNUme3t8zCBiYqrpzuqkdviPJK6bcHGBC9KNhtlXVS9JNv/Sy1tpnp90eYOfYB8OwieHdZ+pCAAAAAAAABsmILgAAAAAAAAZJogsAAAAAAIBBOnzaDRjXUUcd1W55y1tOuxlA76Mf/egXWmtHj1NW/MLsEcMwbGIYhms78ZuIYZg1YhiGTT8ahm2jGB5MouuWt7xlzjvvvGk3A+hV1Z5xy4pfmD1iGIZNDMNwbSd+EzEMs0YMw7DpR8OwbRTDpi4EAAAAAABgkCS6AAAAAAAAGCSJLgAAAAAAAAZJogsAAAAAAIBBkugCAAAAAABgkCS6AAAAAAAAGCSJLgAAAAAAAAZJogsAAAAAAIBBkugCAAAAAABgkCS6AAAAAAAAGKTDJ1VxVR2W5CVJbpzkvCTfTnLfJEckeVpf7NVJLk1yTmvtrZNqCwAAAAAAAPNnkiO6fiLJsekSXPuSPKy19pQk70jy8H45s1/3kAm2AwAAmIKl5ZVU1ZplaXll2s0CAABgjkxsRFeS70ry4dba71fVmUmu6NfvSXLn/vd/7H9evl4FVXVykpOTZGXFATEAAAzJgf17c/wpZ61Zt+fUE6fUGgAAAObRJEd07Uvypf731Ymslf65fUmWN2tHa+201toJrbUTjj766Ik1FAAAAAAAgOGZ5IiudyZ5ZVXdK8m5Sb5UVa9Jcr0kT+/LvKqqHpTkvRNsBwAAAAAAAHNoYomu1to3kjx5ZPXpI4+fNKntAwAAAAAAMN8mOXUhAAAAAAAATIxEFwAAAAAAAIMk0QUAAAAAAMAgSXQBAAAAAAAwSBJdAAAAAAAADJJEFwAAAAAAAIMk0QUAAAAAAMAgSXQBAAAAAAAwSBJdAAAAAAAADJJEFwAwFUvLK6mqNcvS8sq0mwUAAADAgBw+7QYAAIvpwP69Of6Us9as23PqiVNqDQAAAABDZEQXAAAAAAAAgyTRBQAAAAAAwCBJdAEAO8q9twAAAADYLe7RBQDsKPfeAgAAAGC3GNEFAAAAAADAIEl0AQAAAAAAMEgSXQAAAAAAAAySRBcAAAAAAACDJNEFAAAAAADAIEl0AQAA27a0vJKqunJZWl6ZdpMAAABYQIdPuwEAAMDwHNi/N8efctaVj/eceuIUWwMAAMCiMqILAAAAAACAQZLoAgAAAAAAYJAkugAAAAAAABgkiS4AAAAAAAAGSaILAAAAAACAQZLoAgAAAAAAYJAkugAAgCstLa+kqtYsS8sr024WAAAArOvwaTcAAACYHQf2783xp5y1Zt2eU0+cUmsAAABgc0Z0AQAAAAAAMEgSXQAAAAAAAAySRBcAAAAAAACDJNEFAAAAAADAIEl0AQAAAAAAMEgSXQAAAAAAAAySRBcAAAAAAACDdPikKq6q+yR5SZJ/SvK2JN+f5FZJrp3kqUmWkrwsyeVJXt9ae9+k2gIAAAAAAMD8meSIrpbk60mum+TCJHdrrT0jyT8muWeSJyd5aZKTkjxlgu0AAAAAAABgDk1sRFeSD7TW3l9Vt0hyeroEV5LsSbLcL3tba1dU1boVVNXJSU5OkpWVlQk2FQAAAAAAgKGZ2Iiu1toV/a9fSvKVJEf1j1eS7OuX5arasA2ttdNaaye01k44+uijJ9VUAAAAAAAABmiS9+h6eJL7J7lJklckuVtVvTzJEUleneQz6aYuvCzJ6ybVDgAAAAAAAObTxBJdrbV3JnnnqlXnjBS5MMkTJ7V9AAAAAAAA5tvEpi4EAAAAAACASZLoAgAAAAAAYJAkugAAGJSl5ZVU1ZplaXll2s0CAAAApmBi9+gCAIBJOLB/b44/5aw16/aceuKUWjMcS8srObB/75p1xxx7XC7ad8GUWgQAAADXnEQXAAAsAAlCAAAA5pGpCwEAAAAAABgkiS4AAAAAAAAGSaILAJhZS8srqao1y9LyyrSbBQAAAMCMcI8uYENuWs+88tkeDvcUAgAAAGAzEl3AhpxgZl75bAMAALNo9KI8F+QBwNYkugAAAABgBoxelOeCPADYmnt0AQAAAAAAMEgSXQAAAAAAAEzM0vJKqmrNsrS8siN1m7oQmDmjc5In5iUHAAAAABiq0el5k52boleiC5g5k/zSAwAAAABgfpi6EBisSQ53BQAAAABgd42e8x3nfK8RXcBgGfkFADA8pqkGAAA2MnrOd5zzvRJdAAAA7BoXKwEAADvJ1IXA3DuU4a4AADCvTAEOMH98twOLzIguYO4dynBXAIZvdHo0U6MBdIyqA5g/vtuBRSbRBQDAXHKhAwAAAMw/UxcCAAAAAMw50xsC88qILgAAAACAOWd6Q2BeGdEFAAA7wBWyAAAAsPuM6AIAgB3gClkAAAAOxdLySg7s33vl42OOPS4X7btgii0aFokuAACYMaMHOYkDHQAAgHk1euGkiya3Z3CJLgf9AADMO6PDgEPlmBkAgEUzuESXg34AAABYn2NmAAAWzWHTbgAAAAAAAAAcCokuAAAO2dLySqpqzbK0vDLtZgGwA3zHAwAwBIObuhAAgNlhiiyA+eU7fnrca42dMPo58hkCYLvG6ZPMQr9FogsAAGAOOcEJwzX0JOMsnPDi6p+jIX2GAJgN4/RJZqHfItEFAAAwh5zgnC8SBwzJLJzwAgAWh0QXAAAAzDiJAwAAWN9h024AAAAAHIql5ZVU1ZplaXll2s2CuTYad2IOAJi2hR7RZeoHAIDFpj8Iw2aUE+w+06ICALNmoRNdDooAYHtGkwISAgzdNPqDkmsAAABM07wdl0400VVVN0hybpJfTvJdSW6V5NpJnppkKcnLklye5PWttfdNsi0AwDXnCl645lxsBfNv3k4cAAAwfTvZx5y349JJj+g6Jck70t0L7G6ttcdV1TOS3DPJfZK8NMk/JXlLEokuAAAABm/eThwAADB9+pgbO2xSFVfV/ZJ8MsnnkxyZ5OL+qT1Jlvtlb2vtik3qOLmqzquq8y6++OKNigEAAAAAALCAJpboSnLfJD+U5LH9cvN+/UqSff2yXFUbtqG1dlpr7YTW2glHH330BJsKAADAvFpaXklVXbksLa9Mu0kAAMAOmdjUha21FyRJVZ2U5AtJbl9VL09yRJJXJ/lMuqkLL0vyukm1AwAAgGHZ6XtcucckMG3u3QcAkzPpe3SltfaGDZ66MMkTJ719AAAAhsX9B4B5s5Pfa5JmMGxiGHbexBNdAMB0jHaedZwBAGD4XAwAwyaGYedJdAHAnDJNE+wcV10CwM6zfwUAdoJEF7CrHMgAMESuugSAnWf/CgDsBIkuYFc5kAEAAGCWuUATAIZFogsOkXvfAAAAwPxxgSYADMvcJrokIZg0974BYN65mhkAAACYdXOb6JKEAGCIXKjBLNnJq5klzQAAAIBJmNtEFwAMkQs1mFemAAIAAGCeuFh5dkh0AQAAACw4o68BYHtcrDw7Dpt2AwAA4KCl5ZVU1ZXL0vLKtJsEAAvh4Mm61cto4gvYfaP9Y33k6Rjn/+B/BdNjRBcAADPDFXEAw2IUEMBkmQJ8Nozzf/C/gumR6AIAAKbK3PYMic/rWk7qAcDOcyEJbI9EFwAAMFVG8s2HRTkh4/MKANOxKH2NxIUksF0SXQBwDSxSRxsANuOEDDAJ+tuT5f1lSPQ1gI1IdAHANaCjzRA5oQEADIX+9mR5fwEmy/H37hgr0VVVj0pydmvtq1X1K0l+IMkLWmsfnWjrAGBO6NgwS5zQAAAAZol7YDKvHH/vjnFHdL2wtfb2qvqBJPdP8vIkr0zyIxNrGQDMER0bAACmxQlkYFrGvejTPTCBa2LcRNe3+58/nuR1rbXTq+o5E2oTAAAAAFtwAhmYdS76hN23iLMKjZvoalX1uCSPSfLgft11JtMkAAAAALbiBDIAMGoR+weHjVnuGUkemeQPWmufrarbJXnf5JoFAAAAAAAAmxt3RNf1WmsPPfigtfbpqnrXhNoEAAAAAAAAWxp3RNdvrbPuZTvZEAAAAAAAANiOTUd0VdVtk9w+yY2r6oGrnjoyyfUn2TAAAAAAgHm1tLySA/v3rll3zLHH5aJ9F0ypRQDDtNXUhfdIclKSWyR57qr1X03ynAm1CeaGDgsAAAAA6zmwf2+OP+WsNev2nHrilFoDMFybJrpaa29M8saqOqm19obdaRLMDx0WgN3hwgIAAACAxbTViK4kSWvtDVV1myS3Wf03rbU/nVTDAADG5cICAAAAgMV02DiFqurXk/xdkhemm8LwuTF1IQA7ZGl5JVV15bK0vDLtJgEAwESM9n31fwF2lu9ZWDxjjehK8lNJbtNa++okGwNcc6bvYohGR+PMykgc8QQAwE4zEh1gsnzPwuIZN9F1kSQXTM7oyfRrciLdzhx2jngCAAAAgNk2bqLrw1X1x0nOSPJfB1e6RxdDMssjM2Z1NAsAAAAAwDyY5fPDXDPjJrru3v985qp1LYlEF4NhZAbsPh0IAJgPi7BPX4TXCACwyJwfnl9jJbpaa/eddEMAmD86EAAw28ZN7izCPn0RXiMsOgltAJhPYyW6quqB6603dSEAAMBwSe4Ai8R3HgDMp3GnLnzuqt+vm+T7kpwfUxcCMGNcpQnAvLOvAwAAuMohTV1YVXdM8qyJtAhgCpwwmh+u0gRg3tnXAQAAXGXcEV1rtNY+WVXfu9ONgVkg4bGYnDACAAAAABieQ7lH12FJ7t7/hLkj4QHMOgl5AAAAAOgcyj26LkvymSSP3OwPquoOSX42yVFJ/jrJV5LcN8kRSZ7WF3t1kkuTnNNae+v4zYa1nPTdHu8XO2H0c+QztHsk5AEAAGaX8y4Au+uQ7tE15t98KslTq+qwJH+Q5MattUdW1YlJHt4XO7O19t6qensSiS4OmZO+2+P9YieMfo58hgAAAMB5F4DdNtb0g9X5mao6o6reUVVPqaoa4+8ekuSD6UZ0tX71niTL/XLw0obLN/j7k6vqvKo67+KLLx6nqQAAAAAAACyIce+z9Zvppip8d5L/1f9+6lZ/1Fp7T2vtR5I8btXqlST7+mV5s3a01k5rrZ3QWjvh6KOPHrOpAAAAAAAALIJx79F1/yR3a61dliRV9Y4kH03yvI3+oKruk26KwiOS/GmSL1XVa5JcL8nT+2KvqqoHJXnvIbUeAAAAAACAhTVuoqty1dSD6X/fdOrC1to5Sc4ZWX36yOMnjbl9AAAAAACAiVpaXsmB/XvXrDvm2ONy0b4LptQitjJuouvPk5xdVW9Il+Q6qV8HAAAAAAAwFw7s35vjTzlrzbo9p544pdYwjk0TXVV1rXRTDz4vycnppiKsJO9JctrEWwcAAAAAAAAbOGyL51+a5LGttStaa69trf1ka+0RSa6V5Ncm37zZsLS8kqpasywtr0y7WQAAAAAAAAttq6kLH5jk+eusf1WSj6cb6TX3DFUEAAAAAACYPVuN6LqitXb56MrW2hVJrphMk+AqRtMBAAAAAAAb2WpE13Wq6vqttW+sXllVN0x37y6YKKPpAAAAgK0sLa/kwP69a9Ydc+xxuWjfBVNqEQCwW7ZKdL09yRur6smtta8mSVUdmeT3k5wx6cYBAAAAwFZcKMtuGU2qzmNCVeIYGJqtEl2/kuQNSfZX1af7dbdL8p4kL5pcswAAAAAAZstoUnUeE6oSx8DQbJroaq1dluTxVXXbJHdNUknOb6392240DgAOckUZAAAAADBqqxFdSZI+sSW5tQknYAEmyxVlAAAAAFdxTho6YyW62JoTsAAAAAAA7BbnpKFz2LQbAAAAAAAAAIdCogsGYml5JVW1ZllaXpl2swAAAAAAYGpMXQgDYSgyAAAAAACsZUQXAAAAAADbMjr7kJmHgGkxogsAAAAAgG0ZnX3IzEPAtBjRBQAAAAAAwCBJdAEAAAAAADBIEl0AAAAAAAAMkkQXAAAAAAAAgyTRBQAAAAAAwCBJdAEAAAAAADBIEl0AAAAAAAAMkkQXAAAAAAAAgyTRBQAAAAAAwCBJdAEAAAAAADBIEl0AAAAAAAAMkkQXAAAAAAAAgyTRBQAAAAAAwCBJdAEAAAAAADBIEl0AAAAAAAAMkkQXAMAcWVpeSVWtWZaWV6bdLAAAAICJOHzaDWBxLS2v5MD+vWvWHXPscblo3wVTahEADN+B/Xtz/ClnrVm359QTp9QaAAAAgMmS6GJqnIgDAAAAAACuCVMXAgAAAAAAMEgSXQAAAAAAAAySRBcAAAAAAACDNLF7dFXVQ5M8KMnNk/xekjsnuVWSayd5apKlJC9LcnmS17fW3jeptgAAAAAAADB/Jjaiq7X27tbaU5KclOQxSe7WWntGkn9Mcs8kT07y0v75p0yqHQAAAEOxtLySqrpyWVpemXaTAAAAZtrERnSt8sIkr0vyyP7xniTL/bK3tXZFVa37h1V1cpKTk2RlxQEeAAAw3w7s35vjTznrysd7Tj1xiq0BAACYfRMb0VWdU5OcneQjSY7qn1pJsq9flqtqwza01k5rrZ3QWjvh6KOPnlRTAQAAAAAAGKBJjuh6ZpL7JTkyyW2TnF9VL09yROve+dkAABRxSURBVJJXJ/lMuqkLL0s34gsAAAAAAADGNrFEV2vtFUlesUmRC5M8cVLbBwAAAAAAYL5NbOpCAAAAAAAAmCSJLgBgISwtr6Sq1ixLyyvTbhYAMCf0NQAApmOS9+gCAJgZB/bvzfGnnLVm3Z5TT5xSawCAeaOvAQDTs7S8kgP7965Zd8yxx+WifRdMqUXsJokuAAAAAABgsFxwsthMXQgAAAAAAMAgSXQBAAAAAAAwSBJdAAAAAAAADJJEFwAAAAAAAIMk0cWOW1peSVWtWZaWV6bdLAAAAACAhTN6vta5WubN4dNuAPPnwP69Of6Us9as23PqiVNqDQAAAADA4ho9X+tcLfPGiC4AAAAAYENm7wFglhnRBQAAAABsyOw9AMwyI7oAAICZ50pyAAAA1mNEFwAAMPNcSQ4AAMB6jOgCAAAAAABgkCS6AAAAAAAAGCSJLgDWcA8UAAAAFo1jYYDhco8uANZwDxQAAAAWjWNhgOEyogtgG0av8HJ1FwAAAADA9BjRBbANo1d4uboLAAAAAGB6jOgCAAAAAABgkCS6AAAAgIkanQLcNOAAAOwUUxcCAAAAEzU6BXhiGnAAAHaGEV0AAAAAAAAMkkQXAAAAAAAAgyTRBQAAAAAAwCBJdAEAAAAAADBIEl0AAAAAAAAMkkQXAAAAAAAAgyTRBQAAAAAAwCBJdAEAAAAAADBIEl0AAAAAAAAMkkTXLltaXklVrVmWllem3SwAAAAAAIDBOXzaDVg0B/bvzfGnnLVm3Z5TT5xSawAAAAAAAIbLiC4AAAAAAAAGSaILAAAAAACAQZLoAgAAAAAAYJAmdo+uqrp1khckObK19pNV9dgk901yRJKn9cVeneTSJOe01t46qbYAAAAAAAAwfyY2oqu19u+ttSevWvWw1tpTkrwjycP75cx+3UMm1Q4AAAAAAADm025OXdj6n3uSLPfL3n7d5ev9QVWdXFXnVdV5F1988S40EQBYdEvLK6mqNcvS8sq0mwUAAADAOiY2deEmVpLs639fTvKxbJBwa62dluS0JDnhhBPaemUAAHbSgf17c/wpZ61Zt+fUE6fUGgAAAAA2M7ERXVV1s6p6bZK7VtXPJ3l3Vb0mySOTvLNfHtGve++k2gEAAAAATJaZEYBJ8N3COCY2oqu19sUkTx1ZffrI4ydNavsAAAAAwO4wMwIwCb5bGMdu3qMLAAAAAAAAdoxEFwAAAAAAAIMk0cW2jM6Jaj5UAAAAAABgWiZ2jy7m0+icqOZDBQAAAAAApsWILgCAQ2CUMwAAAMD0GdEFAHAIjHIGAAAAmD4jugAAAAAAABgkia4ZNDoVkumQAABg9+mXAwAAzD5TF86g0amQEtMhAdfc0vJKDuzfu2bdMccel4v2XTClFgHbIYZh9+mXAwAAzD6JLoAdNqsno52sg2ETwzCeWd0Pz2q7AAAAhk6iC2CHORkNANMzq/vhWW0XzJrRpLCEMAAAW5HoAgAAAGbCaFJYQhgA5pdZD9gpEl0AAMBCcUANAADTZ9YDdopEFwAAsFAcUMOwSVYD22VaVID5JtEFAAAADIZkNbBdpkUFmG+HTbsBAADzaml5JVW1ZllaXpl2swAAYGr0kWE2iU2GzIguAIAJccU5AACsNU4f2RSlsPscvzJkEl0AAAAAm3DSHXaXE+4AbIdEFwAAAMAmnHQHAJhd7tEFAABwiNzLAAAAYLqM6AKYYaZIAYDZZpQHAEyPY2YAEokugJnm5BkAAACszzEz7L7RBLPkMrNAogsAAACYO07EAcDOG00wSy4zCyS6BszwbAAAAFifE3EAAItBomvADM8GAAAAAAAW2WHTbgAAwKJbWl5JVV25LC2vTLtJAAAAAINgRBcAwJSZWgkAAADg0BjRBQAAAAAAwCBJdJHk6lMmmTYJhkUMAwAAAACLyNSFJLn6lEmJaZNgSMQwAAAAALCIjOgCAAAAAAB2hJmH2G1GdAFMydLySg7s37tm3THHHpeL9l0wpRYBAMBi0ScHgJ1n5iF2m0QXwJTY6QMAwHTpkwMADJ+pCxfA6FBRw0QBAAAAAIB5YETXAhi9Qs3VaQAAAAAAwDwwogsAAAAAAIBBmuqIrqq6QZJXJ7k0yTmttbdOsz0AAAAAAAAMx7RHdD08yZmttackeciU2wIAAAAAAMCATDvRtZxkb//75dNsCAAAAAAAAMNSrbXpbbzqCUm+1Fo7q6re1lp79MjzJyc5uX/4XUn+ZdXTRyX5whib2cly6pr9bS5CXdPY5npljm+tHT1G/amqi5Ps2eX2zUJd09jmItQ1jW3OY11ieLh1TWObi1DXNLY56zE8K691Fra5CHVNY5uLUNd65caO32QuY3hW65rGNhehrmlsc2aOhZN1Y3hWX+s0trkIdU1jm4tQ1zXZpmPh4fyv1DVb25yVutaP4dba1JYkN0jy+iSvSfK4bf7tebtdTl2zv81FqGuW27+dZeivdVa3uQh1Db39s1rXdpdZfR2LUNfQ2z+rdc1D+7ezDP21Drn9s1rX0Ns/q3Vtp9x2liG/J7Na19DbP6t1Db3904pf/6v5qWvo7Z/VunZ6m9tZZvU9mdW6ht7+Wa1r6O3f6feitZbDM0WttUuSPGmabQAAAAAAAGCYpn2PLgAAAAAAADgkQ050nTaFcuqa/W0uQl3T2Oa4dW3H0F/rrG5zEeqaxjYXoa7tmtXXsQh1TWObi1DXNLY56zE8y691yO2f1bqmsc1FqGs75bZjyO/JrNY1jW0uQl3T2KZj4eFscxHqmsY2F6Gund7mdszqezKrdU1jm4tQ1zS2Oat1JUmqn+sQAAAAAAAABmXII7oAAAAAAABYYIdPuwGHoqpukOTVSS5Nck5r7a0blLt1khckObK19pMblHlokgcluXmS32ut/cU6Ze6Q5GeTHJXkr1trr9mibecm+eXW2lkblLlPkpck+ackb2utnbNOmcP6MjdOcl5r7Y0b1HWvJI9L97+8Y2vtR9Yps5LkVUm+kORfW2sv3aCuOyZ5UZIv9q/zzFXPrXkvq+qxSe6b5IgkT2utXbJBuWcleXySJ7bWPrFBmdf3mzksyU+31i7foNzzktw+ydFJntRa+8/1yvXrntxv894b1HV2kj1Jvt5ae84mr/OeSR6d5LIkp7bWLlqnzKlJjkzyg0me31r78w3q+sUkt0py0yTPbK3tW6fMr6b7LF6e5NmttW/0da35nCa5c1/XtZM8tbXW1ilz7ySPTHK31trXV73G1eXemuSBSa6V5OLW2rM22N6P9b9fr39fL801ME4MjxO/67V10jE8Tvz25baM4XHity83tzE8TvxuUNfVYnic+N2g3ERjeJz47bc3iBje7X1wX26sGN6pfXBfbldjeLP47Z/fMobHid8N6pq7GB4nfjeoa6IxPE78blCXGL6qzH2yi/3ovtzgYnhW+9EblBPD2yCG15Qb3LHwBuUGE8O73Y/u63IsvHFdgz0W7ssNLoZntR+9QTkxPMP74PXaOqQYnrV+9AblBhPD+tE71I9urQ1uSfKEJA/uf3/7GOXPHKPMTZP84RZlDhujzK8kOSXJiZuUuXeSs5O8IcltNyjzsP7530nyY2O0/6FJfmaD5+538Lkkb9qkjmcnuVf/+3s2ey+TnNH/PDHJEzZ7z9N9WX3PVv+XJC9PctwY5X4uyV02adutkjxvvf/7qjJ/km6Oz+du8TrPTHJqkl9PcsQW7TorybXGeM8ek+QBG5R5d//zp5I8foPP6euTvLV//IyD/7P1Psv9Z+iG43zmk5yR5LAtyvx2kptt9Xkc4/M6dgyv938c5/VsUOYax3DGiN++3NgxnE3it39+7mM4Y8TvSLkNYzhjxO8G7ZpoDGeM+B39LGcGYzhT2gf35TaN4ezQPrgvt6sxnDHid4PP49ViOGPE73r/m8xhDK/zfl0tfkfK7UoMrxObV4vfDcqJ4V3uR/fPDzaG1ykzE/3oDdomhsdYIoa3FcOZ0X70ev+bDCiG12mXY+ExlzgWnosYXicGZqIfPVJODA9gH7xeWzcoM5MxnBnrR4++7xlYDGeM+N2gXfrRrQ126sLlJHv73y/foTpfmC5LuK6qekiSDyb5603K3C/JJ5N8fottfaC19oB0Xz4v3qDMdyX5cOsymk/bor4keWySP97guf8/yaOr6m+SvG+TOt7cl3tZkpttsb3W/9yT7v9xjVTVd6cL3L2blLlOVb0uXVZ3zwZlDkv3Bfm7W2zyka21k5MsVdX3blLuLumy2x9Kd6XCRm37gSTntz7zv4G/7P8HT033WVrPO6vqlUnulfXf1xcmeV2Si/vH673/m36W1yvXX4nxz621K9YrU1XHVNVbkqwk+doYdW9lyDE8Tvwm24vhzeI3WZAY3kb8JuPF8Fjx2297N2J4nPg9WG6WY3jX4zfZOoZ3eB+c7H4Mbyd+EzE8uu2tYnic+E12L4bHid8ry4nhNXa7H53MQQzPcj+637YYHp8YvrpBHwsnw45hx8Lb5lj46gYdwzPaj07E8FD2wcmwY3ju+9F9GcfCA+hHDzXRtS9XvZnX6DVU59QkZ7fWzt+oXGvtPa0bhrnZh+u+SX4oXZA/pf+Qr1fXwX/el9INs1zPvv75ZIsvz+qGgX6ltfbVDYo8Kd2w1R9NF5Draq39R2vt6Umen25I6ThW+rYesqr6niTPSfI/NivXWru0tfbf0mW9f2KDYrdON4z0N5PcpaoeuEFdB/8H/5Hkhpts9lOttcvS/S9utEm5/5bkjzZ5PumuqPjRdF9UT96gXW9qrT0zyceS/PPB9as/p0k+km7IcrLq/R/3szxarrqhyw9N8ksblWmtHWitPb7f9g9v8TrHMdgYHjN+kzFjeIz4TRYnhseK376ucWJ43PhNJhjD48TvaLkZj+Fdj99krBjeyX1wsssxfIjxm4jhg7aK4S3jt2/XRGN4nPhdr5wYXlPPbvejkzmI4RnvRydieDvE8CpDPxZO5iKGHQtvj2PhVYYewzPcj07E8Ezvg5Phx/Ci9KMTx8IZSD96kPfoSvLOJK+qqgclee9GharqZkl+Lcldq+rnW2u/sU6xZ6YbRnlkVd22tfbadeq5T5KHp/sS+NONttdae0Ff/qQkX2jrZyNTVQ9Pcv8kN0k3T+l63pnkldVlNs/daJu9J6cbPriRP0vyourmMP3cRoWq6pZJfiHJDZK8bOS5Ne9lkndX1WvSzY/59E3KXZhuKOkdqupXW2v/OFLmBemGO56d5BV9mX0b1HWTJNdPN3Tx2ettM11W/FH9+uXW2p9uUNd3J/lGuhj4zU3a/5b+dd4wybM2KPOqdEMn17y365T7ZFW9Nt0X369sUOab6eZ8vTzdHLwHXfk5TXLbJOdX1cvTfSZfPVqmqm7bv1c/nOR3q+qXWmsXrlPuHv3/711JXlNVP9da++ZImTsluV26KyZukI0/s9uxZQyPGb+jr2fiMTxm/Cbjx/BW8ZvMcQyPE78b1HW1GB4nfjcoN+kYHid+15Sb8Rje1X1wX9d9skUM7/A+ONnlGN4sfvvnt4zhceJ3nXJzGcPjxO8G5SYdw+PE72hdYnhtXbvdj04GGMOz2o/eoJwY3h4xvNagjoXXKTeoGJ5CPzpxLDzPx8LJwGJ4VvvRG5QTw7O9Dx59PUOM4ZnpR29QbjAxrB+9M/3oaq1t9jwAAAAAAADMpKFOXQgAAAAAAMCCk+gCAAAAAABgkCS6AAAAAAAAGCSJLgAAAAAAAAZJomvBVdXNqupj/XKgqvavenydHd7WT1fVMRs895aqeuiqx4dX1Zd3cvswb8QvDJsYhmETwzBsYhiGS/zCsIlhJuHwaTeA6WqtfTHJ9yVJVb0oyddba781oc39dJLzkxyYUP3rqqrDW2uX7eY2YTeIXxg2MQzDJoZh2MQwDJf4hWETw0yCEV2sq6p+oar+e//7K6vqL/rf719Vb+h/f0BVfbiqzq+qt1fVDfr1d6+q91fVR6vq7Kq6RVU9Kt0X2Nu3m52vqsOq6neq6hNV9Y9V9ZP9+vtV1btXlXttVT2+/31fVf1iVX0oycN25l2BYRC/MGxiGIZNDMOwiWEYLvELwyaGuSYkutjIuUnu1f9+tyQ3qarDk9wzyQeq6uZJnp/kx1prd0vy8SQ/W1VHJHl5kke01r4/yVuSvKS19vYkH0vyqNba97XWLl1nm/+z/9L5WJLzVq1/ZJI7JrlLkh/vy918jNdwSWvtHq21M7b52mHoxC8MmxiGYRPDMGxiGIZL/MKwiWEOmakL2chHkty9qm6S5OtJ/i3JXdN92bw5yY+kC/a/raokuU6SDya5Q5I7Jfmrfv21kuwbc5s/11p7d9IN70zyhX79PZOc3lq7PMmBqvpgkhOSrPfltNrbx9wuzBvxC8MmhmHYxDAMmxiG4RK/MGximEMm0cW6WmvfqqoLkzwxyYeS/GuSH0uy0lr716q6U5I/a609YfXfVdVdk3y8tXavq1V66GqD9Zdl7ajE6448f8kOtgEGQ/zCsIlhGDYxDMMmhmG4xC8MmxjmmjB1IZs5N8lz+p8fSPL0JB/tn/vbJPeuqlsnSVXdoKpul+STSY6tqh/o11+n/xJKkq8ludEhtuPRVXWtqrpFknukG0q6J8md+m3cNMmPHsqLhDklfmHYxDAMmxiGYRPDMFziF4ZNDHNIjOhiMx9I8twkf99a+2ZVfbtfl9ba56vqyelu5nfwRn6/0Fr7dHU353tFVd0o3Wfst5P8U5LXJ3ldVX0zyf9p745tEAaiIAqum0LU5bqoBQnRBU18Aqd2AATWopkK7oKNnnR3OXgXdc8tyTXJI8kkWWfmlSTL9vnfM1vhv/9+Zfgb9gvdbBi62TB0s2HoZb/QzYb5yjIzZ58BAAAAAAAAPubpQgAAAAAAACoJXQAAAAAAAFQSugAAAAAAAKgkdAEAAAAAAFBJ6AIAAAAAAKCS0AUAAAAAAEAloQsAAAAAAIBKQhcAAAAAAACV3jMYl2gS0mioAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 2160x360 with 7 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(ncols=7, figsize=(30,5))\n",
    "plt.subplots_adjust(wspace=0.05)  #Remove some whitespace between subplots\n",
    "\n",
    "for idx, gp in df.groupby(df_users['created_at'].dt.dayofweek):\n",
    "    ax[int(idx)].set_title(gp['created_at'].dt.day_name().iloc[0])  #Set title to the weekday\n",
    "\n",
    "    (gp.groupby(gp['created_at'].dt.hour).size().rename_axis('Tweet Hour').to_frame('')\n",
    "        .reindex(np.arange(0,24,1)).fillna(0)\n",
    "        .plot(kind='bar', ax=ax[int(idx)], rot=0, ec='k', legend=False))\n",
    "\n",
    "    # Ticks and labels on leftmost only\n",
    "    if idx == 0:\n",
    "        _ = ax[int(idx)].set_ylabel('Counts', fontsize=11)\n",
    "\n",
    "    _ = ax[int(idx)].tick_params(axis='both', which='major', labelsize=7,\n",
    "                            labelleft=(idx == 0), left=(idx == 0))\n",
    "\n",
    "# Consistent bounds between subplots. \n",
    "lb, ub = list(zip(*[axis.get_ylim() for axis in ax]))\n",
    "for axis in ax:\n",
    "    axis.set_ylim(min(lb), max(ub)) \n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Retweets "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(10,7))\n",
    "ax.set_title('Comparison of retweets (samples of 10K tweets)')\n",
    "\n",
    "troll_retweets = np.sum(df_trolls['retweet'].sample(10000))\n",
    "nontroll_retweets = np.sum(df_users['retweet'].sample(10000))\n",
    "\n",
    "ax.bar(['troll'],troll_retweets,label='troll')\n",
    "ax.bar(['non-troll'],nontroll_retweets,label='non-troll')\n",
    "\n",
    "ax.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Links"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "def has_link(text):\n",
    "    return 't.co' in text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_link_percentage(df, text_column_name):\n",
    "    vc = df[text_column_name].aggregate(has_link).value_counts()\n",
    "    return vc[True]/(vc[True] + vc[False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "link_percentage_troll = compute_link_percentage(df_trolls, 'content')\n",
    "link_percentage_non_troll = compute_link_percentage(df_users, 'content')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize=(10,7))\n",
    "ax.set_title('Proportion of tweets with links (samples of 100K tweets)')\n",
    "\n",
    "ax.bar(['troll'],link_percentage_troll,label='troll')\n",
    "ax.bar(['non-troll'],link_percentage_non_troll,label='non-troll')\n",
    "\n",
    "ax.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}