e8_bot_detection_on_twitter/p3 - VADER and RandomForestClassifier.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "c52df017-8c43-41fe-b245-b49e520c504e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import re\n",
    "import seaborn as sns\n",
    "import nltk\n",
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "b30f0b3c-2174-4e88-8a10-8c74d561b67d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package vader_lexicon to\n",
      "[nltk_data]     /home/ashish/nltk_data...\n",
      "[nltk_data]   Package vader_lexicon is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nltk.download('vader_lexicon')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "c8e727ff-a5f9-43ef-8308-98e8cdcfb02e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('input/tweets_of_f234_users_1663839312.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "0c2e0ea4-c9fa-4491-a16c-995cf186d4b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "human    371791\n",
       "bot      116846\n",
       "Name: account_type, dtype: int64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['account_type'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "137d9d8d-cdc7-4470-87af-114eb2d09940",
   "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>userid</th>\n",
       "      <th>account_type</th>\n",
       "      <th>clean_tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can put the \"River\" in \"Screwdrivers\" https...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can't extract the \"Pong\" from \"Sponged\" ht...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>Put the \"Nope\" back in \"Inoperative\" https://t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can't spell \"Artworks\" without \"Two\" https:...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can be the \"She\" in \"Astonished\" https://t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488632</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>A Tribute – Joan Rivers on Fitness http://t.co...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488633</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Live in Orlando area? Want a personal trainer?...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488634</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>If you're trying to get fit  keep working. You...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488635</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Ready for summer? Cheap  cool sunglasses - loo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488636</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Hey all. Stayn' healthy is important especiall...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>488637 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    userid account_type  \\\n",
       "0       787405734442958848          bot   \n",
       "1       787405734442958848          bot   \n",
       "2       787405734442958848          bot   \n",
       "3       787405734442958848          bot   \n",
       "4       787405734442958848          bot   \n",
       "...                    ...          ...   \n",
       "488632          3077047801          bot   \n",
       "488633          3077047801          bot   \n",
       "488634          3077047801          bot   \n",
       "488635          3077047801          bot   \n",
       "488636          3077047801          bot   \n",
       "\n",
       "                                              clean_tweet  \n",
       "0       We can put the \"River\" in \"Screwdrivers\" https...  \n",
       "1       You can't extract the \"Pong\" from \"Sponged\" ht...  \n",
       "2       Put the \"Nope\" back in \"Inoperative\" https://t...  \n",
       "3       We can't spell \"Artworks\" without \"Two\" https:...  \n",
       "4       You can be the \"She\" in \"Astonished\" https://t...  \n",
       "...                                                   ...  \n",
       "488632  A Tribute – Joan Rivers on Fitness http://t.co...  \n",
       "488633  Live in Orlando area? Want a personal trainer?...  \n",
       "488634  If you're trying to get fit  keep working. You...  \n",
       "488635  Ready for summer? Cheap  cool sunglasses - loo...  \n",
       "488636  Hey all. Stayn' healthy is important especiall...  \n",
       "\n",
       "[488637 rows x 3 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "61ce577c-77eb-49d6-8b06-b95e42b62537",
   "metadata": {},
   "outputs": [],
   "source": [
    "userid_accountype = df[['userid', 'account_type']].drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "d18bb17a-a9c8-43f7-af25-5dba93aa4db5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "human    146\n",
       "bot       71\n",
       "Name: account_type, dtype: int64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "userid_accountype['account_type'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "973c793e-0d61-4808-ac44-19c4ea1db23e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['We can put the \"River\" in \"Screwdrivers\" https://t.co/fIxh4ep9W2',\n",
       "       'You can\\'t extract the \"Pong\" from \"Sponged\" https://t.co/EjNM3Oq1tl',\n",
       "       'Put the \"Nope\" back in \"Inoperative\" https://t.co/tlPk66hbPB',\n",
       "       ...,\n",
       "       \"If you're trying to get fit  keep working. You CAN'T fail unless you stop trying. Determination wins every time. RT http://t.co/eNSqWGoCpR\",\n",
       "       'Ready for summer? Cheap  cool sunglasses - looks like the designers; most less than $20 http://t.co/Msu1v22X3c http://t.co/p6pOU2h0MX',\n",
       "       \"Hey all. Stayn' healthy is important especially if you have diabetes. Get tips here: http://t.co/RNvf1I2tt5 #staynhealthy #diabetestips\"],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['clean_tweet'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "76963935-6f50-43f2-85b9-c5fcafdc30ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2min 16s, sys: 1.2 s, total: 2min 17s\n",
      "Wall time: 2min 17s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Wall time: 1min 41s\n",
    "pred = []\n",
    "vader_label = []\n",
    "sid = SentimentIntensityAnalyzer()\n",
    "for sentence in df['clean_tweet'].values:\n",
    "       \n",
    "    ss = sid.polarity_scores(sentence)\n",
    "    \n",
    "    pred.append(ss['compound'])\n",
    "    \n",
    "    if(ss['compound'] < 0.05 and ss['compound'] > -0.05):\n",
    "        vader_label.append('neutral')\n",
    "    elif(ss['compound'] >= 0.05):\n",
    "        vader_label.append('positive')\n",
    "    elif(ss['compound'] <= -0.05):\n",
    "        vader_label.append('negative')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "1c60c0e9-bc0b-490b-8875-3dab820497b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['vader_sentiment'] = pred\n",
    "df['vader_label'] = vader_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "1e2d7744-1d23-422d-a5f8-ee662e560ae5",
   "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>userid</th>\n",
       "      <th>account_type</th>\n",
       "      <th>clean_tweet</th>\n",
       "      <th>vader_sentiment</th>\n",
       "      <th>vader_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can put the \"River\" in \"Screwdrivers\" https...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can't extract the \"Pong\" from \"Sponged\" ht...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>Put the \"Nope\" back in \"Inoperative\" https://t...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can't spell \"Artworks\" without \"Two\" https:...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can be the \"She\" in \"Astonished\" https://t...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488632</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>A Tribute – Joan Rivers on Fitness http://t.co...</td>\n",
       "      <td>0.2732</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488633</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Live in Orlando area? Want a personal trainer?...</td>\n",
       "      <td>0.1680</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488634</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>If you're trying to get fit  keep working. You...</td>\n",
       "      <td>0.8608</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488635</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Ready for summer? Cheap  cool sunglasses - loo...</td>\n",
       "      <td>0.7430</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488636</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Hey all. Stayn' healthy is important especiall...</td>\n",
       "      <td>0.5423</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>488637 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    userid account_type  \\\n",
       "0       787405734442958848          bot   \n",
       "1       787405734442958848          bot   \n",
       "2       787405734442958848          bot   \n",
       "3       787405734442958848          bot   \n",
       "4       787405734442958848          bot   \n",
       "...                    ...          ...   \n",
       "488632          3077047801          bot   \n",
       "488633          3077047801          bot   \n",
       "488634          3077047801          bot   \n",
       "488635          3077047801          bot   \n",
       "488636          3077047801          bot   \n",
       "\n",
       "                                              clean_tweet  vader_sentiment  \\\n",
       "0       We can put the \"River\" in \"Screwdrivers\" https...           0.0000   \n",
       "1       You can't extract the \"Pong\" from \"Sponged\" ht...           0.0000   \n",
       "2       Put the \"Nope\" back in \"Inoperative\" https://t...           0.0000   \n",
       "3       We can't spell \"Artworks\" without \"Two\" https:...           0.0000   \n",
       "4       You can be the \"She\" in \"Astonished\" https://t...           0.0000   \n",
       "...                                                   ...              ...   \n",
       "488632  A Tribute – Joan Rivers on Fitness http://t.co...           0.2732   \n",
       "488633  Live in Orlando area? Want a personal trainer?...           0.1680   \n",
       "488634  If you're trying to get fit  keep working. You...           0.8608   \n",
       "488635  Ready for summer? Cheap  cool sunglasses - loo...           0.7430   \n",
       "488636  Hey all. Stayn' healthy is important especiall...           0.5423   \n",
       "\n",
       "       vader_label  \n",
       "0          neutral  \n",
       "1          neutral  \n",
       "2          neutral  \n",
       "3          neutral  \n",
       "4          neutral  \n",
       "...            ...  \n",
       "488632    positive  \n",
       "488633    positive  \n",
       "488634    positive  \n",
       "488635    positive  \n",
       "488636    positive  \n",
       "\n",
       "[488637 rows x 5 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "e540eee1-4219-46d0-b0a4-f10e5d2b1dfa",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('tweets_of_f24_users_with_vader_sentiment.csv', index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9ccbe38c-c5b7-4bc0-adc6-37f5798fe446",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_mean_sentiment = df.groupby(['userid', 'vader_label']).mean().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f98e31ad-62eb-46c6-8da8-c8b700321802",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_mean_sentiment.rename({'vader_sentiment': 'mean_sentiment'}, axis='columns', inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "76a69752-6ca9-4c84-87b1-a3fa4d8da365",
   "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>userid</th>\n",
       "      <th>vader_label</th>\n",
       "      <th>mean_sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>418</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.431048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>418</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>418</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.565365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>586</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.461686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>586</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>635</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>636</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.509822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.344304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>638</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>neutral</td>\n",
       "      <td>-0.000113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>639</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.609808</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>640 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  userid vader_label  mean_sentiment\n",
       "0                    418    negative       -0.431048\n",
       "1                    418     neutral        0.000060\n",
       "2                    418    positive        0.565365\n",
       "3                    586    negative       -0.461686\n",
       "4                    586     neutral        0.000057\n",
       "..                   ...         ...             ...\n",
       "635  1014793793353199617     neutral        0.000009\n",
       "636  1014793793353199617    positive        0.509822\n",
       "637  1034115201979826178    negative       -0.344304\n",
       "638  1034115201979826178     neutral       -0.000113\n",
       "639  1034115201979826178    positive        0.609808\n",
       "\n",
       "[640 rows x 3 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean_sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8b63a1d7-8aff-4456-a931-619e94f3805c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "814a76ef-d96a-458d-babb-bc524f05ae9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_var_sentiment = df.groupby(['userid']).var().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6b0ebf52-ae22-43be-a2bd-a2841873549b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_var_sentiment.rename({'vader_sentiment': 'variance_sentiment'}, axis='columns', inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "46bd42a8-1dca-4298-8c07-9e0c9433dbcf",
   "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>userid</th>\n",
       "      <th>variance_sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>418</td>\n",
       "      <td>0.175295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>586</td>\n",
       "      <td>0.206516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2622731</td>\n",
       "      <td>0.151959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5768282</td>\n",
       "      <td>0.002442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>14157134</td>\n",
       "      <td>0.200867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>212</th>\n",
       "      <td>969324483185119232</td>\n",
       "      <td>0.165845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213</th>\n",
       "      <td>982628478284595200</td>\n",
       "      <td>0.204360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>214</th>\n",
       "      <td>1006905571994472448</td>\n",
       "      <td>0.174320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>215</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>0.161225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>216</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>0.156157</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>217 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  userid  variance_sentiment\n",
       "0                    418            0.175295\n",
       "1                    586            0.206516\n",
       "2                2622731            0.151959\n",
       "3                5768282            0.002442\n",
       "4               14157134            0.200867\n",
       "..                   ...                 ...\n",
       "212   969324483185119232            0.165845\n",
       "213   982628478284595200            0.204360\n",
       "214  1006905571994472448            0.174320\n",
       "215  1014793793353199617            0.161225\n",
       "216  1034115201979826178            0.156157\n",
       "\n",
       "[217 rows x 2 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_var_sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "31cb053e-08b4-4b49-8f35-9eb61f80359c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_mean_var = df_mean_sentiment.merge(df_var_sentiment, on=['userid'], how = 'inner')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2d71a5e2-7fec-4678-aba3-1a41b06c27fd",
   "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>userid</th>\n",
       "      <th>vader_label</th>\n",
       "      <th>mean_sentiment</th>\n",
       "      <th>variance_sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>418</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.431048</td>\n",
       "      <td>0.175295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>418</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000060</td>\n",
       "      <td>0.175295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>418</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.565365</td>\n",
       "      <td>0.175295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>586</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.461686</td>\n",
       "      <td>0.206516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>586</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000057</td>\n",
       "      <td>0.206516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>635</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.161225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>636</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.509822</td>\n",
       "      <td>0.161225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.344304</td>\n",
       "      <td>0.156157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>638</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>neutral</td>\n",
       "      <td>-0.000113</td>\n",
       "      <td>0.156157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>639</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.609808</td>\n",
       "      <td>0.156157</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>640 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  userid vader_label  mean_sentiment  variance_sentiment\n",
       "0                    418    negative       -0.431048            0.175295\n",
       "1                    418     neutral        0.000060            0.175295\n",
       "2                    418    positive        0.565365            0.175295\n",
       "3                    586    negative       -0.461686            0.206516\n",
       "4                    586     neutral        0.000057            0.206516\n",
       "..                   ...         ...             ...                 ...\n",
       "635  1014793793353199617     neutral        0.000009            0.161225\n",
       "636  1014793793353199617    positive        0.509822            0.161225\n",
       "637  1034115201979826178    negative       -0.344304            0.156157\n",
       "638  1034115201979826178     neutral       -0.000113            0.156157\n",
       "639  1034115201979826178    positive        0.609808            0.156157\n",
       "\n",
       "[640 rows x 4 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean_var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "37917b8e-1177-460a-afa6-f009166a065b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_mean_var_w_label = df_mean_var.merge(userid_accountype, on = 'userid', how = 'inner')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "db6bade4-4dd1-4bd1-b505-9dc1aaeae033",
   "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>userid</th>\n",
       "      <th>vader_label</th>\n",
       "      <th>mean_sentiment</th>\n",
       "      <th>variance_sentiment</th>\n",
       "      <th>account_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>418</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.431048</td>\n",
       "      <td>0.175295</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>418</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000060</td>\n",
       "      <td>0.175295</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>418</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.565365</td>\n",
       "      <td>0.175295</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>586</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.461686</td>\n",
       "      <td>0.206516</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>586</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000057</td>\n",
       "      <td>0.206516</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>635</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0.000009</td>\n",
       "      <td>0.161225</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>636</th>\n",
       "      <td>1014793793353199617</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.509822</td>\n",
       "      <td>0.161225</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>negative</td>\n",
       "      <td>-0.344304</td>\n",
       "      <td>0.156157</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>638</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>neutral</td>\n",
       "      <td>-0.000113</td>\n",
       "      <td>0.156157</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>639</th>\n",
       "      <td>1034115201979826178</td>\n",
       "      <td>positive</td>\n",
       "      <td>0.609808</td>\n",
       "      <td>0.156157</td>\n",
       "      <td>human</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>640 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  userid vader_label  mean_sentiment  variance_sentiment  \\\n",
       "0                    418    negative       -0.431048            0.175295   \n",
       "1                    418     neutral        0.000060            0.175295   \n",
       "2                    418    positive        0.565365            0.175295   \n",
       "3                    586    negative       -0.461686            0.206516   \n",
       "4                    586     neutral        0.000057            0.206516   \n",
       "..                   ...         ...             ...                 ...   \n",
       "635  1014793793353199617     neutral        0.000009            0.161225   \n",
       "636  1014793793353199617    positive        0.509822            0.161225   \n",
       "637  1034115201979826178    negative       -0.344304            0.156157   \n",
       "638  1034115201979826178     neutral       -0.000113            0.156157   \n",
       "639  1034115201979826178    positive        0.609808            0.156157   \n",
       "\n",
       "    account_type  \n",
       "0          human  \n",
       "1          human  \n",
       "2          human  \n",
       "3          human  \n",
       "4          human  \n",
       "..           ...  \n",
       "635        human  \n",
       "636        human  \n",
       "637        human  \n",
       "638        human  \n",
       "639        human  \n",
       "\n",
       "[640 rows x 5 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean_var_w_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "dbc23123-9783-432c-9a82-bc9f1d54682c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "human    435\n",
       "bot      205\n",
       "Name: account_type, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean_var_w_label['account_type'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "7361ac53-ac42-4b07-aa03-118c456e3158",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='mean_sentiment', ylabel='variance_sentiment'>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data = df_mean_var_w_label, x = \"mean_sentiment\", y = \"variance_sentiment\", hue = \"account_type\", style = \"account_type\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "7c9b2a47-d69a-4b0a-b906-df199b63cb8d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='account_type', ylabel='mean_sentiment'>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data = df_mean_var_w_label, x = \"account_type\", y = \"mean_sentiment\", hue = \"account_type\", style = \"account_type\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "64de3737-5478-4d28-81bd-6795b6e73364",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='account_type', ylabel='variance_sentiment'>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAGwCAYAAABB4NqyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAA9hAAAPYQGoP6dpAABvnElEQVR4nO3dd3iUVdrH8e/MJJPeE1IgEEqkN2mCICoogiKouIAiAi6KawVR110FxF1RdJVVXFlQmrtiRZa1oG4EFaVIR0FKKKEkpJHeZ+b9Y14GIwmSkHkGkt/nuuYizznPnOceUObmVJPD4XAgIiIi0oCYPR2AiIiIiNGUAImIiEiDowRIREREGhwlQCIiItLgKAESERGRBkcJkIiIiDQ4SoBERESkwfHydAAXIrvdzvHjxwkKCsJkMnk6HBERETkHDoeD/Px84uLiMJvP3sejBKgKx48fJz4+3tNhiIiISC0cOXKEJk2anPUeJUBVCAoKApy/gcHBwR6ORkRERM5FXl4e8fHxru/xs1ECVIVTw17BwcFKgERERC4y5zJ9RZOgRUREpMFRAiQiIiINjhIgERERaXCUAImIiEiDowRIREREGhwlQCIiItLgKAESERGRBkcJkIiIiDQ4SoBERESkwVECJCIiIsYoyvnVdbZHwgAdhWGYQ5mF7DyWy7f7MogO9uX6TrFEBlqJDPT1dGgiIiLul5MCW/8NXW6DsGaQcxR++ghaXweRiYaHowTIAAczC7hz4Q+kZBe5yuau3s8LIzpxVetGRAT6eDA6ERERN8s5Ap8+Bns/g4PfwO+WwA8L4Ls5sHkh3LHCmRQZSENgbpaZX8ILq/ZUSn4AHA54/MOdZBeWeSgyERERg1is0Hao8+eU72FuD2fyA9D0MjiHw0vrmhIgN8struDzXSeqrLPZHaw/kGVwRCIiIgYLioZWA+GGvzuvS3Kcv7a/Ga78E4Q2NTwkJUBuVmG3Y7M7qq3PLS43MBoREREPsZVD9v7KZXnHwWH3SDhKgNzMz9tCYqPAaut7t4wwMBoREREPOJkCP8yH7191Xkd3cA57HVkPH02Ck4cND0kJkJs1jQjgqRvaVTm82adlBI2CtApMRETqOROne3q63AZ3LIdh//j/uT/Vj5K4NSSHw+GZJ1/A8vLyCAkJITc3l+Dg4PNuLyO/mOSMImav+pmtR3II87dyx2XNGNGtCfHh/nUQsYiIyAUu9ygc2wJxXZxzfvJPQNoOiLykzlaA1eT7WwlQFeo6ATrl2Mkiymx2TJiIC/XF6mWps7ZFREQueLYKsHhVf32eavL9rX2ADNQ4TL09IiLSgP062anD5KemNAdIREREGhwlQCIiItLgKAESERGRBkcJkIiIiDQ4SoBERESkwVECJCIiIg2OEiARERFpcJQAiYiISIOjBEhEREQaHCVAIiIi0uBcEAnQa6+9RkJCAr6+vvTq1YuNGzdWe++CBQvo168fYWFhhIWFMXDgwDPuHzduHCaTqdLruuuuc/fHEBERkYuExxOgd999lylTpjB9+nS2bNlC586dGTRoEOnp6VXev2bNGkaPHs3q1atZt24d8fHxXHvttRw7dqzSfddddx2pqamu17Jly4z4OCIiInIR8Php8L169aJHjx7MnTsXALvdTnx8PA888AB//OMff/P9NpuNsLAw5s6dy9ixYwFnD1BOTg4rVqyoVUzuOg1eRERE3Kcm398e7QEqKytj8+bNDBw40FVmNpsZOHAg69atO6c2ioqKKC8vJzw8vFL5mjVraNSoEa1bt+bee+8lKyur2jZKS0vJy8ur9BIREZH6y6MJUGZmJjabjejo6Erl0dHRpKWlnVMbjz/+OHFxcZWSqOuuu46lS5eSlJTE888/z9dff83gwYOx2WxVtjFr1ixCQkJcr/j4+Np/KBEREbngeXk6gPPx3HPP8c4777BmzRp8fX1d5aNGjXL93LFjRzp16kTLli1Zs2YNAwYMOKOdJ554gilTpriu8/LylASJiIjUYx7tAYqMjMRisXDixIlK5SdOnCAmJuas733xxRd57rnn+OKLL+jUqdNZ723RogWRkZHs37+/ynofHx+Cg4MrvURERKT+8mgCZLVa6datG0lJSa4yu91OUlISvXv3rvZ9s2fP5plnnmHVqlV07979N59z9OhRsrKyiI2NrZO4RURE5OLm8WXwU6ZMYcGCBSxZsoTdu3dz7733UlhYyPjx4wEYO3YsTzzxhOv+559/nqeeeoqFCxeSkJBAWloaaWlpFBQUAFBQUMCjjz7K+vXrOXToEElJSQwbNoxWrVoxaNAgj3xGERERubB4fA7QyJEjycjIYNq0aaSlpdGlSxdWrVrlmhidkpKC2Xw6T3v99dcpKytjxIgRldqZPn06M2bMwGKxsGPHDpYsWUJOTg5xcXFce+21PPPMM/j4+Bj62UREROTC5PF9gC5E2gdIRETk4nPR7AMkIiIi4glKgERERKTBUQIkIiIiDY7HJ0E3JCVlFZwsKsfXaibMXxOyRUREPEUJkAFsNhuHsor5YMtR1idnERloZVyf5iRE+tM4zN/T4YmIiDQ4SoAMsDe9kJH/XEdeSYWr7Mvd6Uzq34JxvROICfXzYHQiIiINj+YAuVlqbjEz/7urUvJzyryvD5BTXO6BqERERBo2JUBuVlBSwboDWdXWf7s/08BoREREBJQAud1v7TJps2sfShEREaMpAXKzQKuFS5uGVVvfr1WkgdGIiIgIaBK028WF+TNjaDt+N38dD17diq5NwygsreCZj3fRp1UUof7eng5RRESkwdFZYFWo67PAysrKOZpbyqof0/guOZOoQB/GXNaM6GAf4sMD6iBiERERqcn3t3qADHAwu4TRC9aTXVjmKlux7TiPXHsJt3ZrQkyIlsGLiIgYSXOA3OxEbjF//WRXpeTnlL99sZfc4jOXx4uIiIh7KQFys/zSirMudf9Oy+BFREQMpwTIzewOONssq5Jym3HBiIiICKAEyO0CrBY6NQmptr7fJVEGRiMiIiKgBMjtGof5M31oO6yWM3+rh3WJI1zL4EVERAynVWAGaB0dxH/uv5y5X+1j8+EcIgKtTLg8gZ7Nw3UavIiIiAcoATJAoK83oT7l/GlIW4rLbVhMJgJ9LAT7WT0dmoiISIOkBMgAR7KL+NNHO/l23+kVX8F+Xiy4ozudGgfj56NhMBERESNpDpCb5ReXseDbA5WSH4C84gomLP6B1LxSD0UmIiJioIpyKMyA0oLTZflpkH/CI+EoAXKzjIIy3t90tMq6wjIb24/mGhyRiIiIwSrKIW07/OMyOL4Fyoshcy+8eQ3s/9IjSZASIDcrs9kpPsteP0eziwyMRkRExAOKM+GTKVCYCf8eATs/gH/dAjkp8NljgPHHkioBcjMfLzMxwb7V1neKr36PIBERkXohKBZuWQjhLaCiFFbe70x+vP1hzIfga/x3oRIgN2sa5sdDAxOrrGsRGUBChE6DFxGRBiCyFdy8oHLZ5Q9DdEfwNv5QcCVAbmaxWOjTMoLpQ9sR9v+bHppMcFXrKN64szt+VpOHIxQRETFA5l54f1zlsrV/c84J+uXEaIMoAXKz4znFPPbBDkzAe/f0ZuX9l7PqoX7cd1Urfr/kB7IKdBq8iIjUc/lp8MEEyD3iHPYasej0cNjbv4PyQsND0j5AblZYWsGGg9lsOJhdZf23+zJoGxtscFQiIiIG8o+CG+fCO6NhxEKIvRRiOsGykXDlnwCL4SEpAXKz35rXfraT4kVEROoFi8U51+fur8EnCLx9nHOCxn0CJi8IjDQ8JA2BuVmgj4UeCWHV1vdLNP4PXURExHAWCwQ2qjzhOSjGI8kPKAFyu7hQf6bd0B5/65nde3f2bkaov84DExERMZqGwAxwSXQg/7nvcpauP8wPB7MJD7ByV9/mtI4OIi7U+KV/IiIiDZ0SIAP4eFtIjA7ij9deQnZxOT5eFhqdZXNEERERcS8lQAYK8LMS4KchLxEREU/THCARERFpcNQDZJCUrEKO55aw6VA2McG+dG0WRnSQD4G+3p4OTUREpMFRAmSAlKxC7l+2lR1Hc11lPl5m5o3pRo+EMCVBIiIiBtMQmJvlFJXx6lf7KyU/AKUVdib9azMn8ko9FJmIiEjDpQTIzbIKyli5/XiVdaUVdjannDQ4IhEREVEC5GbldjulFfZq60/klRgYjYiIiIASILfz9TLTJKz6zQ67Nav+mAwRERFxDyVAbpYQGchj17Wpsq59XDBNtBO0iIiI4ZQAGeDS+FBeGd3V1RNktZgZcWlj/nH7pTSNCPBwdCIiIg2PlsEbID2/hNaNAlk6oScFpRX4elnwt1r4aFMKw7o1pWmEv6dDFBERaVCUALlZSlYhYf5Wkn5OZ8XWY/RICOdYThGl5Xam39gem636CdIiIiLiHkqA3CzE18KavZlccUkUHRuHsPVIDp2ahNCxSQjrk7Po0yrS0yGKiIg0OEqA3Cy7uIIuTcN46J1tbDuS4yr39XbuBF1ms3kuOBERkQZKk6DdzNfbzD9WJ1dKfgBKyu3c89ZmzCb9EYiIiBhN375uVlRmZ8W2Y1XWlVbY2aadoEVERAynBMjNym1n3wk6VTtBi4iIGE4JkJv5WH5jJ+im2glaRETEaBdEAvTaa6+RkJCAr68vvXr1YuPGjdXeu2DBAvr160dYWBhhYWEMHDjwjPsdDgfTpk0jNjYWPz8/Bg4cyL59+9z9MaoUE+TD1GtbV1nXNjaIxmdJjkRERMQ9PJ4Avfvuu0yZMoXp06ezZcsWOnfuzKBBg0hPT6/y/jVr1jB69GhWr17NunXriI+P59prr+XYsdPzbGbPns0rr7zCvHnz2LBhAwEBAQwaNIiSEuOHm06WlNMyKoCXfteZuBBfALwtJoZ3ieMft3cDh/YBEhERMZrJ4XA4PBlAr1696NGjB3PnzgXAbrcTHx/PAw88wB//+MfffL/NZiMsLIy5c+cyduxYHA4HcXFxPPLII0ydOhWA3NxcoqOjWbx4MaNGjfrNNvPy8ggJCSE3N5fg4ODz+nz70/O59uVvuPfKlgzr3JgKux1vi5m03GKmvLedB65OZGyfhPN6hoiIiNTs+9uj+wCVlZWxefNmnnjiCVeZ2Wxm4MCBrFu37pzaKCoqory8nPDwcAAOHjxIWloaAwcOdN0TEhJCr169WLduXZUJUGlpKaWlpa7rvLy82n6kM9jsDuwOeG11Mq+tTj6jPq+kvM6eJSIiIufGo0NgmZmZ2Gw2oqOjK5VHR0eTlpZ2Tm08/vjjxMXFuRKeU++rSZuzZs0iJCTE9YqPj6/pR6mWn7eF1tFB1dZrJ2gRERHjeXwO0Pl47rnneOedd/joo4/w9fWtdTtPPPEEubm5rteRI0fqLMYAqxfThrbDbDqz7srWUUQEWOvsWSIiInJuPJoARUZGYrFYOHHiRKXyEydOEBMTc9b3vvjiizz33HN88cUXdOrUyVV+6n01adPHx4fg4OBKr7qSWVhKel4J70/qTe8WEfh4mYkN8eWP17VhxtD2fPlTap09S0RERM6NRxMgq9VKt27dSEpKcpXZ7XaSkpLo3bt3te+bPXs2zzzzDKtWraJ79+6V6po3b05MTEylNvPy8tiwYcNZ23QXi9nElPe3M+uT3Tx2XWs+ebAvi8f3AOxc+eIaLBaL4TGJiIg0dB4/DHXKlCnceeeddO/enZ49ezJnzhwKCwsZP348AGPHjqVx48bMmjULgOeff55p06bx9ttvk5CQ4JrXExgYSGBgICaTiYcffpi//OUvJCYm0rx5c5566ini4uIYPny44Z/P3+pF/8Qo1uzN4KZ/fF+pzmSCPi01B0hERMRoHk+ARo4cSUZGBtOmTSMtLY0uXbqwatUq1yTmlJQUzObTHVWvv/46ZWVljBgxolI706dPZ8aMGQA89thjFBYWcvfdd5OTk0Pfvn1ZtWrVec0Tqq0Km40/Xd+Wn47ncU27aAa0iSK7qJy/fPITf7gyEZtd+wCJiIgYzeP7AF2I6nIfoD1peXy7N50BbWPYeCibb/ZmEh3sw82XNsHfauHTnancf3ViHUUuIiLScF00+wA1BEFWL65sE8Ntb2wgNff0TtQLvzvEzGHtGdopzoPRiYiINEwX9TL4i4G3t5nnV+2ulPycMmPlT5TZNAQmIiJiNCVAbpZbXEHS7qrPNbM74PvkTIMjEhERESVAbmaz27GfZZZVQWmFccGIiIgIoATI7fysFtrGnuUoDC2DFxERMZwSIDeLDvbhqevbYaniLIyr20QR4a+jMERERIymBMjNjp8s4WRhGR9M6s3lrSLw87bQJMyPPw1pw1M3tOPoySJPhygiItLgaBm8m5Xa7Ny3bCudm4QwfWh7gvy8sNkcfPlTGle9+DWTr7mEPolRng5TRESkQVEC5GbeZjO+3ma2H83l5te/P6O+SaifB6ISERFp2DQE5maRgVZuubRxlXV+3ha6NA01NiARERFPyD8BGT9D7jHndWkBpO+CnBSPhKMEyM1KK+zcfUVLeiaEVSoPsFp4487u+FnOnBwtIiJSr+SfgH1fwGu9YOf7UJAOx7fA/Cvhq794JAnSEJib5ZaUU1Zh47lbOpFbXM6Oo7lEBfnQOjqIAB8TPxzKIS48wNNhioiIuI+tDPaucv78v+nO5GfvKqgohaM/gAeOJa1VD9CECRPIz88/o7ywsJAJEyacd1D1SYiPhR1H8rj6b19z99JN/Hgsl8XfH2TAS19z24If6BAX4ukQRURE3Cs0Hq57Dlpf77ze9R9n8hPeAm5/H8KaGR5SrRKgJUuWUFxcfEZ5cXExS5cuPe+g6pOCMjt/T9oHQJCvF+EBVqICfQA4kFlIipbBi4hIQ+AfBt3HVS5LvBZ8qt8s2J1qNASWl5eHw+HA4XCQn5+Pr6+vq85ms/Hpp5/SqFGjOg/yYlZms+NvNfPl5CvIKihj0+FsrmrdiMnXtObrvelsO5JL/0v0eyYiIvVYaQEc2wzvjqlcvmEeBMVCx1shpOoFQ+5SowQoNDQUk8mEyWTikksuOaPeZDLx9NNP11lw9YG3xczC8T15cNlWdhzNdZX7eJmZN6YbjYJ8PBidiIiIAUrz4KuZp4e9xq6Ezx6HPZ/Ad3Og0+8MD6lGCdDq1atxOBxcffXVfPjhh4SHh7vqrFYrzZo1Iy4urs6DvJgF+1iY/fneSskPOFeHTfrXZj59sJ+HIhMRETFIcBzc/AZ8OR0GTnPOCRr8PPiFQs+7ISDa8JBqlAD1798fgIMHDxIfH4/ZrFX0vyWvxMbK7cerrCutsLM55SQtGwUaHJWIiIjBwpvDkNkQFOO8Do2Hq590Jj8Wi+Hh1GoZfLNmzcjJyWHjxo2kp6djt9sr1Y8dO7ZOgqsPyu12Sivs1dafyCsxMBoREREPOpX8nBLsuVGjWiVA//3vf7n99tspKCggODgYk+n0Zn4mk0kJ0C/4eJlpEubH0ZNnrpoD6NYsrMpyERERcZ9ajWE98sgjTJgwgYKCAnJycjh58qTrlZ2dXdcxXtQCrBYeu64NAHEhvtzWsyl9W0UA0D4umJhg37O9XURERNzA5HDUfPvFgIAAdu7cSYsWLdwRk8fl5eUREhJCbm4uwcHB59VWcnoBecUlhPr7kllYxo6jOTQK8qVdbDCBVhM/HM7hhs7GLv0TERGpj2ry/V2rIbBBgwaxadOmepsA1aVAHwteFj8mv7udLSknXeX+Vgvz7uhG16YaAhMRETFarRKg66+/nkcffZRdu3bRsWNHvL29K9XfeOONdRJcfWA2mZj3dXKl5AegqMzG3Us38ckDWgYvIiJitFolQBMnTgRg5syZZ9SZTCZsNtv5RVWP5JVUsHzLsSrrSsrtbDuSo2XwIiIiBqtVAvTrZe9SvXLb2ZfBH8+tenWYiIiIuM9572RYUqJ9bM7Gx8tM41C/auu7xocaF4yIiIgAtUyAbDYbzzzzDI0bNyYwMJADBw4A8NRTT/Hmm2/WaYAXu4hAbx659sxz0wBaRwcRH159ciQiIiLuUasE6K9//SuLFy9m9uzZWK1WV3mHDh1444036iy4+iA9v5ym4f68MKKTa88fL7OJ6zvGMO+Obvx4LPc3WhAREZG6Vqs5QEuXLmX+/PkMGDCASZMmuco7d+7Mzz//XGfB1Qc2u517/rWZT++/nO7NelFcbsPby0yg1cKKzUep+MUu2iIiImKMWiVAx44do1WrVmeU2+12ysvLzzuo+sTHy8KH9/Zhyvs7OHqyiGvbx3D0ZDHf7c9k/thuRAdpJ2gRERGj1SoBateuHd9++y3NmjWrVP7BBx/QtWvXOgmsvgjyNfPWuiNMH9oOP6sXJeU2rBYTJtMlJO1OJ+oSJUAiIiJGq1UCNG3aNO68806OHTuG3W5n+fLl7Nmzh6VLl/Lxxx/XdYwXtZxiG7f3akJeiZ33Nh1hXXIWkUE+3Nm7GVe2jqawtMLTIYqIiDQ4tToLDODbb79l5syZbN++nYKCAi699FKmTZvGtddeW9cxGq4uzwJLzy8iq6CCkfPXkVdcOdm576qWjOweT9OIgPN6hoiIiBhwFhhAv379+PLLL2v79gajrNzBMx/vOiP5AXhtdTI3dIrzQFQiIiINW60ToFMKCgrO2Bn6fHtN6pPicjvfJ2dVW792fyZtY/X7JSIiYqRa7QN08OBBrr/+egICAggJCSEsLIywsDBCQ0MJC9Pp5r/0W+OLFbZajUCKiIjIeahVD9CYMWNwOBwsXLiQ6OhoTNrLplr+VguXNg0lp6iMWTd3IsTPG5vdwYptx1jw7UGuSIz0dIgiIiINTq0mQQcGBrJ582Zat27tjpg8ri4nQZ/IKaLEZievuIJXvtrH5sM5hAdYGd8ngctbRWAxmYjXJGgREZHzVpPv71oNgfXo0YMjR47UKriGxt/Xwom8Mm5+/Xu+3JVOdmEZ+9ML+POKH3n+8z2/OUQmIiIida9WQ2BvvPEGkyZN4tixY3To0AFvb+9K9Z06daqT4OqDkwUVzPz4J8qrmOvz2c407u3fkqYeiEtERKQhq1UClJGRQXJyMuPHj3eVmUwmHA4HJpMJm81WZwFe7IorbPx4LK/a+rX7MunUJNS4gERERKR2CdCECRPo2rUry5Yt0yTo32A2mzCZoLqZVn5Wi7EBiYiIeEJRDlQUgdkLAhtBRTkUpAImCI03PJxaJUCHDx9m5cqVVR6IKpX5epnpnxjFmr0ZNAv34+o20Rw9WcSXu9MxmaBPS60CExGReq4oB46sg/fugBvnQuIgOHkQ3hoO3cZDj98bngTVKgG6+uqr2b59uxKgc+BwOHjqxrb82daW7MIyth7JoVOTEB4f3IaTBaXU8iQSERGRi0d5IXw/F2zlsGIS9PoDbPsXlOTC9mXQc6LhIdUqARo6dCiTJ09m586ddOzY8YxJ0DfeeGOdBFcf+HhZcFTYeeiDbWw7kuMq9/U2M29MN+JCz3szbhERkQtbSGMY/g9YfjccWQ/rX3OWB0bD2BUQ0sTwkGq1D5DZXP3q+fowCbou9wHKzC/ihc/38+6mM7cN8PEy8+lD/WgZFXhezxAREbngVZTDsY2waMjpsn5Toc/94Fc3p0i4fR8gu91e7etiT37qWm6JnRXbjlVZV1phZ+vhkwZHJCIiYrCKcjixE5aNrly+9m+w93MoSDc8pFolQHLuym12Sivs1dan5pUYGI2IiIgHFGfCfx9yzvkJbAT3fAvxlzmXSH88ufql0m50zhNQXnnlFe6++258fX155ZVXznrvgw8+eN6B1RdWi5kmYX4cPVlcZX23pjo8VkRE6rmgWBjxJqz4Awx9BaLbwc3z4b8PwhWPgTXI8JDOeQ5Q8+bN2bRpExERETRv3rz6Bk0mDhw4UGcBekJdzgHKKijh231ZPPzutjPq2sUG84/bLyUhUmeBiYhIA5CfBkExp6/zjoFPGPj410nzNfn+rtUk6PquLhOgwpJyUnNL2HEslxc/38Px3BK8LSau7xjLwwMvwcsLmoQqARIRETlfbp8EPXPmTIqKis4oLy4uZubMmTVu77XXXiMhIQFfX1969erFxo0bq733p59+4pZbbiEhIQGTycScOXPOuGfGjBmYTKZKrzZt2tQ4rrqQU1yOr7eZrvGhLJnQk08e7MunD/bjnita4GUCH+2iLSIiYrhaJUBPP/00BQUFZ5QXFRXx9NNP16itd999lylTpjB9+nS2bNlC586dGTRoEOnpVc8ILyoqokWLFjz33HPExMRUeQ9A+/btSU1Ndb3Wrl1bo7jqipfFxM8nCrjqb19zzcvfcP0ra7nm5W8Y/MpaJr+3nfyy6idIi4iIiHvUKgE6dejpr23fvp3w8PAatfXSSy8xceJExo8fT7t27Zg3bx7+/v4sXLiwyvt79OjBCy+8wKhRo/Dx8am2XS8vL2JiYlyvyEjPHDlRVGbj2U92V1n3w+GTZBaUGRyRiIiI1Ggb4rCwMNeQ0iWXXFIpCbLZbBQUFDBp0qRzbq+srIzNmzfzxBNPuMrMZjMDBw5k3bp1NQntDPv27SMuLg5fX1969+7NrFmzaNq0aZX3lpaWUlpa6rrOy6v+9PaaKquwcyCzsNr6DQez6Nm8ZkmjiIiInJ8aJUBz5szB4XAwYcIEnn76aUJCQlx1VquVhIQEevfufc7tZWZmYrPZiI6OrlQeHR3Nzz//XJPQKunVqxeLFy+mdevWpKam8vTTT9OvXz9+/PFHgoLOXGo3a9asGg/dnSuL2YyX2USFveq55uEBVrc8V0RERKpXowTozjvvBJxL4vv06XPGGWAXisGDB7t+7tSpE7169aJZs2a899573HXXXWfc/8QTTzBlyhTXdV5eHvHxdXMqrb/VzJAOsazccfyMOi+ziV7NI+rkOSIiInLuanUSZ//+/bHb7ezdu5f09HTs9soTea+44opzaicyMhKLxcKJEycqlZ84ceKsE5xrKjQ0lEsuuYT9+/dXWe/j43PW+UTno7zczpRrL+Gn1FySM04PhVnMJl4e2QWTdiEQERExXK0SoPXr13Pbbbdx+PBhfr2NUE0OQ7VarXTr1o2kpCSGDx8OOM8ZS0pK4v77769NaFUqKCggOTmZO+64o87aPFeNgnz4NjmLf97RjYOZhXy/P4voYB8GtI0mOaMAL4uWwYuISANRkgu+p6fPUHQS/D1zIkKtEqBJkybRvXt3PvnkE2JjY6tcEXaupkyZwp133kn37t3p2bMnc+bMobCwkPHjxwMwduxYGjduzKxZswDnxOldu3a5fj527Bjbtm0jMDCQVq1aATB16lSGDh1Ks2bNOH78ONOnT8disTB69Oiqg3Cj1PxSpv3nJ9LySujTMoIhHWNIzS3hmpe/AWDphJ40i9Rp8CIiUs/lpMD2d6HT7yCsGeQchV3/hUuugchWhodTqwRo3759fPDBB66E43yMHDmSjIwMpk2bRlpaGl26dGHVqlWuidEpKSmYzadX6x8/fpyuXbu6rl988UVefPFF+vfvz5o1awA4evQoo0ePJisri6ioKPr27cv69euJioo673hrqsxmJ+3/Dzz9PjmL75OzKtVvO5LDFZcYH5eIiIhhco7A0S3QfRwc3Qre/pCyDjre5Dwe4+RhZ1JkoFolQL169WL//v11kgAB3H///dUOeZ1Kak5JSEg4Y9jt19555506iasueJvN+HlbKC6velgwPqxuzj8RERG5YFn8IDAKXukKt7zpTHii28M/r4DLH4a2Qw0PqVYJ0AMPPMAjjzxCWloaHTt2PGM1WKdOneokuPog1M/CiG5NeGv94TPq/K0WOsWHVPEuERGResRRBl88CaX58M5tMGA6rJsLBSfg6+eg/XDDQ6rVYai/HJJyNWQyuXaIPtdJ0BequjwMdU9aHn5WL/744Y5Kw19BPl4sGNudFlG+NArWYagiIlKP5WZART68exuk/+J0BN8QuGMFhDSBwEbn/ZiafH/Xqgfo4MGDtQqsIWoa5sf7W47x1PVtKbXZ2Z2aT2SgleaRAXy5K42YkDhPhygiIuJmDnAAQ1+FNweeLr7sD+AfDuUlhkdUqwSoWTNjJypdzFLzSpnzv31kF5YR6ufN5a0iSMkuYucx53EbrRoFkRCpHiAREanHLBYoKYF3b69c/s0LEHcpRHc0PKRaHYYK8NZbb3H55ZcTFxfH4cPO+S1z5szhP//5T50FVx+UVdjJLnQeeDp5YCLj+iTw6KDWtI52Ln3flZrvyfBERETcz1EOyyc65/z4hsCYj6BRO7BXwHt3OBMkg9UqAXr99deZMmUKQ4YMIScnxzXnJzQ0lDlz5tRlfBc9b4uZCX2akjSlP9lF5bzwxV7e+eEIT9/YgeX39iaxkXp/RESknvMOgJvnQ0RL55yf9N0w8l8Q0wluXQy2CsNDqtUk6Hbt2vHss88yfPhwgoKC2L59Oy1atODHH3/kyiuvJDMz0x2xGqYuJ0HnF5VwNLeMUfPXk1tcXqnugatbcWu3JjSNUBIkIiL1WFEOlOaClw+snQMbXodGHeD2d50rw3xDIPj858TW5Pu7Vj1ABw8erLQZ4Sk+Pj4UFhZW8Y6GK7fUxl8+2XVG8gPw6lf7KapmfyAREZF6w1YCa2aBrRyKsmD8KgiMAFuZczdoc62mJJ+XWiVAzZs3Z9u2bWeUr1q1irZt255vTPVKcZmd7/ZnVVv/7b6Lu7dMRETkN5l94YqpsOJe6PcIlOTD0Fdg1RPQ4grA+DlAtUq5pkyZwn333UdJSQkOh4ONGzeybNkyZs2axRtvvFHXMV7Ufmt8scKm0+BFRKSesxfDBxMgdTv86ya44yP4z/1w6Fvn64EthodUqwTo97//PX5+fjz55JMUFRVx2223ERcXx9///ndGjRpV1zFe1Hy8zHSND2XrkZwq6/slRhobkIiIiOG8Ycjf4K3hkHccXut1umrQc87VYAar9TL422+/nX379lFQUEBaWhpHjx7lrrvuqsvY6gWrxcy0oe3w8TJjsUDHxsFEBVoBGNGtMQFW47v9REREDFVRCA4H3PcDRLZ2lpkscP8maDUA7MbPh61VD1BxcTEOhwN/f3/8/f3JyMhgzpw5tGvXjmuvvbauY7yoBfp44WWGLydfQWGZjQMZBYQHWIkO9sXf20yJhsBERKTes0BQtHPjw9HL4N0xcOsS2LwEutwGgbGGR1SrBGjYsGHcfPPNTJo0iZycHHr27InVaiUzM5OXXnqJe++9t67jvGj5W82E+Fn5yye7+WLXCVd5ZKCV+Xd0Jz7c14PRiYiIGMAMLP8DHF4LRzc69wJKehq2/Rt2vge/TwLCDA+pxrZs2UK/fv0A+OCDD4iJieHw4cMsXbqUV155pU4DvNhlF5bz1rrDlZIfgMyCMsYt2khesZbBi4hIPWcyQ+/7wGyBEz/CnPbO5Aeg80jn8JjBapUAFRUVERQUBMAXX3zBzTffjNls5rLLLnMdiyFOeSUVvL0xpdq6n47nGhyRiIiIwWzlENcFbll4+hqg61jocQ/nMSW51mr1xFatWrFixQqOHDnC559/7pr3k56eft47J9c35XY7hWXV9/IczCoyMBoREREPKS+GfZ9XLju+BWyl/PamMXWvVgnQtGnTmDp1KgkJCfTq1YvevXsDzt6gqnaIbsisFjNRgT7V1ndqHGJgNCIiIh7y7Yuw7W3nzwn9Tg+HvTcWMH46SK0SoBEjRpCSksKmTZtYtWqVq3zAgAG8/PLLruujR49it9vPP8qLWLCfF/dd1bLKuvhwPxIidQ6YiIjUc2YviGrj/LnPg3DTP+F3/3ImQSFNAJPhIdXqMNRzFRwczLZt22jRooW7HuEWdXkY6s9peTgcDpJ2pzPv6wMUlDo3e+rVPJy/3tSRH4/mMPzSJnURtoiIyIUpL9V56GlxNvhHgJevc0isJBcCIp2JUGjT839MDb6/3Xr6mBtzq4uG1WLGBHRrFsZHf+hDQWkFvt4W7HYHdpudNjGBng5RRETEvWxlsOZZCG8B3/4NEgfBga+gx91w5DsYscTwkIw/frWBCff34o21h5i7OvmMujB/bz68t48HohIRETGQyQQDpkHSTOf1qcnQ+7+EUcvA4m14SMavO2tgsosqWLqu6q0BThaV83NansERiYiIGMzLF7z8oePvKpc37w/eflw0y+Dl3JXZ7OSVVH/I2770QgOjERER8QAvX0j/Cd4f67w2/f+k5x8WwNZ/geMiWQV2rkwm42d1X2isFjPhAdZq69vFBhkYjYiIiAeU5cPXzzs3QIxqDZN3Q/ubnXUb/4knVoG5NQHSJGgItFq4+4qqV8HFBPvSMkqToEVEpJ6z2Z1L3zuNhJFvO88Bu+rP0P0uuOM/YKl+vzx3Oa9J0Pv37yc5OZkrrrgCPz8/HA5HpV6fXbt2ERcXd95BXswKy2xc3yGG9jFBRIf6UV5hx9vLTEFJOSE+3niZlCSKiEg9ZwJ+XAFXPAY7P4QB0+HQWmcCtO8L6HATEGloSLXqAcrKymLgwIFccsklDBkyhNTUVADuuusuHnnkEdd98fHxWCyWuon0IuVntZBXUkFafinv/JDC0ZPFbDqUzVvrDmLxMntg828RERGDhcZDxxGQtRfaD4OX2oDVHwpPOJOfsATDQ6pVAjR58mS8vLxISUnB39/fVT5y5MhKO0MLeJlN5BaVcWmzMNrFBPPpzlQOZhby+ytacSS7ELtD86RERKSeKy2Ak4cgLx3+0ctZ9s5tkLYT9v4Pco4YHlKthsC++OILPv/8c5o0qbyDcWJiok6D/5WikgqiQ/wY++ZGjuUUu8rfWHuQvw7vQGyonwejExERMUBpHvxvGhzb4twMcex/4NPH4Mtp4BsKbYcYHlKteoAKCwsr9fyckp2djY+P8ROZLmQWLzOzP/+5UvID4HDAkyt+RPPERUSk3vMJhlvehDY3wG3vwf+egcHPOydF37ECvIxfEFSrBKhfv34sXbrUdW0ymbDb7cyePZurrrqqzoKrD4rL7fxvd3qVdXYHrEvOMjgiERERgxWfhO9egWv+Autfh+v+CpsWQr9H4OfPoCzH8JBqNQQ2e/ZsBgwYwKZNmygrK+Oxxx7jp59+Ijs7m++++66uY7yo2ex2bPbqu3nySsoNjEZERMQDvHyhWR+Y29U5BLLzPefhqN/Nga5juWj2AerQoQN79+6lb9++DBs2jMLCQm6++Wa2bt1Ky5Yt6zrGi5qvt4XW0dVvdnh5K2OX/YmIiBjOXg6BjeDGV53XpfnOXzveCpc/AGbjjyat9RNDQkL485//XJex1Evhfl5MG9qOO97cwK87gvpfEklEYPW7RIuIiNQLJqtzqfveX60UP3kYzN5gM35CbK16gBYtWsT7779/Rvn777/PkiXGH2l/QTOZ8bLAe/f05rIW4VgtZmKCfXn8utbMuLED5R74QxcRETFURQFs+Kdz/g9AbBfneWBHN8JH9wBlhodUqwRo1qxZREaeOXTTqFEjnn322fMOqj5JLyjlzoU/8Nynu/nj4LZ8+lBflkzogRm46sU1bEvJ8XSIIiIi7mUynz7uoutYuO1duGm+MwnywPAX1HIILCUlhebNm59R3qxZM1JSUs47qPqk3GanpNzOppQchr925gTxo79aHi8iIlLvWIOh2zho2gsiEuHldjD6PRizHMJagMn4JKhWPUCNGjVix44dZ5Rv376diIiI8w6qPvGxmIkL8a22vlvTUOOCERER8YTyQvjPHyDnKLx3h/M0+B/fh9QdsGykJxaB1S4BGj16NA8++CCrV6/GZrNhs9n46quveOihhxg1alRdx3hRC/LzZsq1l1RZl9gokCZh2glaRETqOYcJhr0Gadth5L+cPT79/wgFJ2DEQrBU31HgLrXqc3rmmWc4dOgQAwYMwMvL2YTdbmfs2LGaA/QrWQWlNI8M4LmbO/K3L/eSkV+KxWzi2nbRPDaoNesOZJEQafwOmCIiIobx8oaULdDnQXh3DOSkQHkx3L0Gjm6CllcbHpLJ4aj9YQx79+5l+/bt+Pn50bFjR5o1a1aXsXlMXl4eISEh5ObmEhwcfF5tpWQVMn7xJm65NI6B7WIoq7BjtZjJLixlynvb+fvorvRICK+jyEVERC5QBRmQexQWXHm67KZ/QrO+ENqk2rfVRE2+v88rAaqv6jIBAvjq53QmLP7hjPJezcN5YUQnmkYEnPczRERELmg5KZCf5twV+p/9YPjrEHkJBERBWN10oNTk+7tWQ2A2m43FixeTlJREeno6dru9Uv1XX31Vm2brpZOFpQT7WHjrrp4899nP/HQ8j2BfL27r1ZRRPZpi1z5AIiJS3+UchsyDENMOUnfClN3OhCi4MWQnO++poyToXNUqAXrooYdYvHgx119/PR06dMBk8sD07YtEdmE5o9/YwPO3dOTV0V0pq7DjZTFhNsEt875j6rVtSGikOUAiIlKPWfzBL8TZ8zNiEWTuh+BYWDwErnwCQo2fQlOrIbDIyEiWLl3KkCFD3BGTx9XlENietDxOFpaRnl/K7M/3cPRkMVaLmaGdY3ng6kTW7stgTO+EuglcRETkQpSfBv8eAWk7wdsPbpwLq/8K2QfA2x8e2gaB0ef9mJp8f9dqGbzVaqVVq1a1Cq6hiQr0Ib2gjAff2cbRk85ND8tsdj7ccoz73t5C38QoD0coIiLiZkExzuXuoU2dq78+vMuZ/Hj5wm3vgXf1h4a7S60SoEceeYS///3vaP70b8srKeeFz3+usu6n43kcz9VO0CIiUs8V5YDdAXf9D3xCTpf/fjWEt4TiLMNDqtUcoLVr17J69Wo+++wz2rdvj7e3d6X65cuX10lw9UFJhZ0j2dUnOZsPn6RPyzPPVRMREak37OXg7Qs73nHu/TP/SrjjIziyDlpcCT5hhodUqwQoNDSUm266qa5jqZe8zWZ8vMyUVtirrI8ONn73SxEREUOVF8OX02HXR5C8Gh7YBJsWwZpnITIRbv8QMHZPvFolQIsWLarrOOqtiEArN3aO4/3NR8+o8/Ey062p8VmviIiIoSze0Lwf9JviXAn2eh8ozIAJX8BPKy6es8Dk3IX6W3ng6lZ0bBxSqdzHy8zrY7oRHezjochEREQMEhwLidc4z/56aIcz+fnDRqgohcsmeWQZfK3Pn//ggw947733SElJoaysrFLdli1bzjuw+sJms7H18En+9rvOZBeWsflwNtFBvnRpGsoPh7LJLAgg0Nf7txsSERG5WJ08DFvfgm9egG7j4ZG9cGgtLL8LEvo5l8UbvBFirXqAXnnlFcaPH090dDRbt26lZ8+eREREcODAAQYPHlzj9l577TUSEhLw9fWlV69ebNy4sdp7f/rpJ2655RYSEhIwmUzMmTPnvNt0p5STxfz1s5+59uVv+NPyHRSUVrB6TzoDX/qGJ5b/yKGsQo/EJSIiYhwHlOY5f9y8CN4Z5Ux+HA7n/CCMX1VeqwToH//4B/Pnz+fVV1/FarXy2GOP8eWXX/Lggw+Sm5tbo7beffddpkyZwvTp09myZQudO3dm0KBBpKenV3l/UVERLVq04LnnniMmJqZO2nSn0go76fmlABzILOL1NQf4ZGeaq3770Zr9fomIiFx8TNDjbuh5t/Py2BZn8tP4Urh5PjgshkdUqwQoJSWFPn36AODn50d+fj4Ad9xxB8uWLatRWy+99BITJ05k/PjxtGvXjnnz5uHv78/ChQurvL9Hjx688MILjBo1Ch+fqufP1LTN0tJS8vLyKr3qitVixt9q4fXbL+XLyVfw4b29+eTBvnz6YF+6Nw0lIcK/zp4lIiJyQQprBl5+ENulcnlEKzB7QXi84SHVKgGKiYkhOzsbgKZNm7J+/XoADh48WKPNEcvKyti8eTMDBw48HZDZzMCBA1m3bl1tQqtVm7NmzSIkJMT1io+vuz+IAKuFjx/oy0/Hcxk6dy23vL6O619ZyxPLd/LsLZ3okWDssj8RERHD5aU69/xZeZ/z2hrg/HXHe/Dt35wHoxqsVgnQ1VdfzcqVKwEYP348kydP5pprrmHkyJE12h8oMzMTm81GdHTl8z+io6NJS0ur5l113+YTTzxBbm6u63XkyJFaPbsqFouJpJ/Tmbs6mZLy03sBbT+ay7iFGymtsNXZs0RERC5I9grYvsw57NWkBzy4DXpNctYlf+WRkGq1Cmz+/PnY7c4v8/vuu4+IiAi+//57brzxRu655546DdAIPj4+1Q6nna+84grmrUmusu54bgkHMgppHqnT4EVEpD4zwZAXnROgu41zFvWYCAGR0PZGsBi/JUytEiCz2YzZfLrzaNSoUYwaNarG7URGRmKxWDhx4kSl8hMnTlQ7wdkTbZ6Psgo7WYVl1db/eDyPAW3P/wRcERGRC1ZoE+dS+EvHgckKgY2cewC1HeY8DT7I+O/Bcx4C27Fjh6vXZ8eOHWd9nSur1Uq3bt1ISkpyldntdpKSkujdu3cNPoZ72zwf3l5mgnyqzzNbRgUYGI2IiIiHhDUDv1AIa+K8Do0H/3BncuQB59wD1KVLF9LS0mjUqBFdunTBZDJVOeHZZDJhs537vJYpU6Zw55130r17d3r27MmcOXMoLCxk/PjxAIwdO5bGjRsza9YswDnJedeuXa6fjx07xrZt2wgMDKRVq1bn1KaRIgK8GHNZM95ad4gXbu1M4zA/Kmx2Xknaz7ajObSLDTY8JhEREY/w/9XCnwDPHQZ+zgnQwYMHiYqKcv1cV0aOHElGRgbTpk0jLS2NLl26sGrVKtck5pSUlErDbcePH6dr166u6xdffJEXX3yR/v37s2bNmnNq00hZhRWM7RPPTZc2Zun3h9hwMJvIQB/GX57AzGHtyS0qNTwmERGRhs7kqMm6daC8vJx77rmHp556iubNm7srLo/Ky8sjJCSE3NxcgoPPr4fmUGYheSXljPzneorLK/eM3d6rKRMuT6Blo6DzeoaIiIjU7Pu7xsvgvb29+fDDD2sdXEPjZTEx87+7zkh+AP69IYWSCnsV7xIRERF3qtU+QMOHD2fFihV1HEr9VFRmY9Phk9XWf7sv08BoREREPKSiHAozobTgdFl+GuSfqP49blSrZfCJiYnMnDmT7777jm7duhEQUHkl04MPPlgnwTUEZpOnIxAREXGzinJI2wHLfgcjFkOT7pB7BP41Avo/Dq0GGr4UvsZzgICzzv0xmUwcOHDgvILytLqcA3Q0u5CpH+xg/YHsKus/e6gvbWNDzusZIiIiF7T8VPj375xJkJcPXP8SfP288wgMawA8sLVOEqCafH/XKgGq7+oyASoqreBgViGj/rme/NKKSnW/79ucsX2a0TRcewGJiEg9l7kf3r4Vsn/RSeLtD3csdx6S6u133o+oyfd3rYbA5Nz5+3gR6GPho/su590fjrDxYBYRAT7ceXkCsSE+mE0aAxMRkQYgshXcvADeGHC67PKHIbpjnSQ/NVXrBOjo0aOsXLmSlJQUysoqH/Xw0ksvnXdg9cXR7EImv7uDH4+d5K83dWRwx2jKKhy8uOpnNqXk8MkDfWkS5ukoRURE3CxzL3zwqw2J1/4NmvWGuEvBx9hzMWuVACUlJXHjjTfSokULfv75Zzp06MChQ4dwOBxceumldR3jRa2o3M6WFOcqsEc/2HlG/bf7M2nfWHOARESkHstPgw8mOOf8ePvDjXNh9V+cw2Fv/w4e2m54AlSrZfBPPPEEU6dOZefOnfj6+vLhhx9y5MgR+vfvz6233lrXMdZrFi0DExGR+s4/ypn0BMU65/y0uQFufx8iWjrLa5eOnJdaPXH37t2MHTsWAC8vL4qLiwkMDGTmzJk8//zzdRrgxS7AaqF3i4hq6/u18tw5KCIiIoawWCCsJdzzLUS0Bm8fCG0O4z6F+MsgMMrwkGqVAAUEBLjm/cTGxpKcnOyqy8zUxn6/5HDYefKGtgT7njnaOKl/C7y9jM96RUREDFWcA0fXw5z2cHCN8zr9R/jHZbB1KeQeNTykWs0Buuyyy1i7di1t27ZlyJAhPPLII+zcuZPly5dz2WWX1XWMF7WCMjvv/3CE5X/ow/Itx1h/IJvIQCt39kkg0MeLT3ek8sCARE+HKSIi4j7lRfDt36CiFD68y7n6a/MiKD4Jm96ES+80PKRa7QN04MABCgoK6NSpE4WFhTzyyCN8//33JCYm8tJLL9GsWTN3xGqYutwH6EBGAde/spYyu41nbuxA65ggSspsPLfqZ3Yey+PlkV24qWvjOopcRETkAnXykHMV2LEtp8sCIuGO/0BMhzp5hNv3AXr22WcZM2YM4BwOmzdvXm2aaRCiAq3c2r0JS9cd5k8f/VipLsBqoZNWgImISEMQHA+DZsHCQafLuk2AkHiPhFOrCSgZGRlcd911xMfH8+ijj7J9+/a6jqveCPKzMrFfC/olVp7sHOznxcJxPWgSZvzmTyIiIoay2ZxzfpaNqlz+7YuQnAQFGYaHVOujME6ePMn777/P22+/zbfffkubNm24/fbbue2220hISKjjMI1Vl0NgAGk5RRSX28kvreDntDwiA31pFu5HoI8X0SFKgEREpJ7LT4N/j4C0neAfAbe9B5896hwO8/Zz7gMUaOxZYLVeghQWFsbdd9/NmjVrOHz4MOPGjeOtt96iVatWtW2y3tp2NJfZn+/By2zi8paRtIz0Y0vKSe5/ezMpWYWeDk9ERMS9gmLgljedZ36NXek8DX7EImjWB0a/C17+hod03meBlZeXs2nTJjZs2MChQ4eIjjb2OPsL3aGsQvy8zdzeqymPfbiDH4/lEezrxW29mvLCrV05UVBC0wgdhioiIvVcVGu47V1nMgQQluBMiqxB4BtkeDi17gFavXo1EydOJDo6mnHjxhEcHMzHH3/M0aPGr+W/kPmYTdgcJsa8uZEfj+UBkFdSwbyvD/D4hzuI8PfxcIQiIiIGOZX8nBIc55HkB2rZA9S4cWOys7O57rrrmD9/PkOHDsXHR1/kVSm1OfjrJ7uqrNtwMJuswjJaGL8BpoiISINWqwRoxowZ3HrrrYSGhtZxOPVPaYWN5Izq5/msP5BFj4RwAyMSERGRWiVAEydOrOs46i2L2YyX2USFverFduEBVoMjEhERER1E5WZh/t4Mal/1xHAvs4lezas/KFVERETcQwmQm0UE+vDooDY0j6y80stsghdv7UxEgLeHIhMREWm4znsZvPy2hMgAFo3vwZ7UfL5LziQ62Jdr20UTHmAlLECTx0VERIymBMggCREBJEQEMKhDzG/fLCIiIm6lITARERFpcNQDZJC03GLySyo4llNMsK834QFWEiK1A7SIiIgnKAEywJHsIl76ci8rth3j1NGzjUP9eH3MpbSPDcJisXg2QBERkQZGQ2BuVlRawTs/pPDR1tPJD8CxnGLuXLiRlJPFngtORESkgVIC5GZpeSUs/f5wlXUni8r56XiewRGJiIiIEiA3K7PZyS+tqLb+wFmOyRARERH3UALkZlaLmYizHHfRNtYzp+CKiIg0ZEqA3KxJmB93X9GiyrrYEF8SGykBEhERMZoSIDezelm4vmMs9/ZviY/X6d/ujo1DWDKhp5bCi4iIeIDJ4XBUfUx5A5aXl0dISAi5ubkEBwfXSZu5xWVk5peRU1SGr9VCiK83TcL966RtERERqdn3t3qADJJTVE5yRgErtx/nm70ZFJXbyC4o9XRYIiIixsk/ATbb6eu841CU45FQlAAZ4FBmIeMX/cDdb21mybrDPL9qD9fN+YZv92dyslBJkIiINAAndsGS6yFthzMJyj4Ay0bD0Q0eSYKUALlZVkEpL36xhwOZlZe72x3wyHvbySos91BkIiIiBslPg+UTIXMfvDUMDn8L746B1G3w3liwlxkekhIgN8spKmfVj2lV1lXYHWw4mGVwRCIiIgYLioGb50NAFJTkwtJhcOInMHvBrYvBXP12Me6iBMjNKux2KuzVzzPPLjQ+6xURETFUUTaYzDDq7crl/aZCZGsoKzA8JCVAbubjZaZlVPVL3S9rHmFgNCIiIh5QXgoWH/j44crlG16Hkhyw+BkekhIgNwuwWvjTkLZV1vVMCCMqyMfgiERERAxmApbfdXrYa+DTENjIORz21nBwaA5QvRPibyU62JelE3rSPs65J0Gwrxf3XNGCF27tjMXD8YmIiLidTwgMnAnWQBj5L4i/DO5Y4UyCrngMsBsekpfhT2xgrF4WQv292XU8lxdGdML6/7tBH0gvoLjMRpvYutloUURE5IJl8YaQeHhwG+z7ApaNgo63wt3fQGk+mIxPR5QAGaBJmD/XtIsmq7CMk4Xl+FottI0LJj5cx2CIiEgDUJwJ74+FrGTwCYI/rIdPH4W53QEHPLDF8JCUABkkLMCHsADN9xERkQYoKBZGLISVD8L1L0GjNjDsNfh0KvSbAr4hhoeks8Cq4I6zwACyC0rJLS7Hy2ImXueAiYhIQ5OXCsGxp69zj4F/OHjXzSqwmnx/qwfIAAUl5aRkF/Ha6v1sPpxDRKCVcX0SuKxFuIbBRESk4fhl8gMQ0tgzcaAEyBB7TuQzev4GymzOWe5peSU8+sEOhnWJ47FBrWkcpt4gERERI2kZvJsdPVnE0//d5Up+fuk/246TXaSzwERERIymBMjNisps7DiaW239t3szDIxGRETEQ3KPwsGvIeeI87owCw5/DycPeyScCyIBeu2110hISMDX15devXqxcePGs97//vvv06ZNG3x9fenYsSOffvpppfpx48ZhMpkqva677jp3foRqmU1gMlVf7+utrRBFRKSeyzkC296GJTfCurlQcAIOr4XF18PHkz2SBHk8AXr33XeZMmUK06dPZ8uWLXTu3JlBgwaRnp5e5f3ff/89o0eP5q677mLr1q0MHz6c4cOH8+OPP1a677rrriM1NdX1WrZsmREf5wz+Vgv9WkVWW395K50FJiIi9Z0D8o45f9wwD94bC++PA4cdirOd9Qbz+DL4Xr160aNHD+bOnQuA3W4nPj6eBx54gD/+8Y9n3D9y5EgKCwv5+OOPXWWXXXYZXbp0Yd68eYCzBygnJ4cVK1bUKqa6XAZ/KDOf0gqY93Uyv+/bHB9vZ8655dBJTuSXMqBtI9rFGb//gYiIiKFyjsK3L8DmxafLYrvArUsgPKFOHlGT72+P9gCVlZWxefNmBg4c6Cozm80MHDiQdevWVfmedevWVbofYNCgQWfcv2bNGho1akTr1q259957ycrKqjaO0tJS8vLyKr3qSrkN8kvKubVbE/700U4GvvQNt7y+jsMnixjWNY7NKSfr7FkiIiIXLG8/SOhbuSy6nfOYDA/waAKUmZmJzWYjOjq6Unl0dDRpaWlVvictLe0377/uuutYunQpSUlJPP/883z99dcMHjwYm81WZZuzZs0iJCTE9YqPjz/PT3ZaZKAPBaUV3P7mBrb//2To3OJyXludzGMf7OCKxKg6e5aIiMgFqTALDn8Hy+92Xnv9/8kI296G7185PTHaQB6fA+QOo0aN4sYbb6Rjx44MHz6cjz/+mB9++IE1a9ZUef8TTzxBbm6u63XkSN39QeSXlPPMx7upaqBx/YFsMvJL6+xZIiIiF6SKYtj0hnPOT2wXePhH6DbOWbd7JZ6YA+TRjRAjIyOxWCycOHGiUvmJEyeIiYmp8j0xMTE1uh+gRYsWREZGsn//fgYMGHBGvY+PDz4+7jmnq7jcRnJGQbX16w9k0T0h3C3PFhERuSCENIGb3gSTw5kEBTaCK/8MVz4BtjIIbWp4SB7tAbJarXTr1o2kpCRXmd1uJykpid69e1f5nt69e1e6H+DLL7+s9n6Ao0ePkpWVRWxsbLX3uIvFbMbLXP06+FB/q4HRiIiIeIDNBvlH4Z9XQPYBKC+F0jxYNBiObIAC4/fE8/gQ2JQpU1iwYAFLlixh9+7d3HvvvRQWFjJ+/HgAxo4dyxNPPOG6/6GHHmLVqlX87W9/4+eff2bGjBls2rSJ+++/H4CCggIeffRR1q9fz6FDh0hKSmLYsGG0atWKQYMGGf75wvy8GNQ+uso6i9nEZS20DF5EROq5ogxY+QDkp8K/boaf/wtv3+pMhlY+AJx5WoK7efwssJEjR5KRkcG0adNIS0ujS5curFq1yjXROSUlBbP5dJ7Wp08f3n77bZ588kn+9Kc/kZiYyIoVK+jQoQMAFouFHTt2sGTJEnJycoiLi+Paa6/lmWeecdsw19lEBPkydVAbfjyex+GsIle52QQvjOhEuL9nZr+L1JTD4aCioqLaxQTiGd7e3lgs2lBVLnBBMTBiIfzrFshJgQ/vcpZ7+cJt74O38QeDe3wfoAtRXe4DdMqhzEJ2pebx3f5MooN9GNQ+lshAKxGBxidlIjVVVlZGamoqRUVFv32zGMpkMtGkSRMCAwM9HYrIbzu6Cd74xVzcq/4Ml/0BfOrmv9+afH8rAaqCOxIgkYuV3W5n3759WCwWoqKisFqtmM52vosYxuFwkJGRQVFREYmJieoJkgtb5v7Tw16nePvDHcudK8O8/c77ETX5/vb4EJiIXNjKyspcO7T7+/t7Ohz5laioKA4dOkR5ebkSILlw5afCB+OdyY+XL9zwEqx5zjkc9q9b4IGtdZIA1YQSIAOVVdjILijDz2ohRKu/5CLzy7l4cuFQb5xcFPwiYeir4O0LJQUQ0w6a9IDyEijO8UhISoAMYLPZOJRdzEdbj7IuOZvIQCt39kkgITyAuDBjM14RERHDeXmDf5hz1+fe94PV39kTtPXf0HkUBFW9WtqtIRn+xAZob3ohI+evI6+4wlX2+U8n+MOVLbmjdzNiQ5QEiYhIPZZ7FP5zPxz6Bo5shDEfwtcvwNalsOMduOt/EFp3x1CdC/Vpu1lqbjEzP95VKfk55R9rksktLvdAVCIiIgby9ode94DZAmk74OX2zuQHoMMtgPFDuUqA3KygpIJ1ydWfRP/tvkwDoxGRqiQkJDBnzpxzvn/x4sWEhoa6LR6Reqes0LnXz81vOK9tZc5fu46FS8eBw/j9xZQAudlv7TFgs2sXAhERqecsVghPgOSvKpenbgOLF5iM3xRYCZCb+XmbubRpKK2iAnj37stY9VA/PnmgLxP7NQegX6tID0cocn5WrVpF3759CQ0NJSIightuuIHk5GRX/dGjRxk9ejTh4eEEBATQvXt3NmzY4Kr/73//S48ePfD19SUyMpKbbrrJVXfy5EnGjh1LWFgY/v7+DB48mH379rnqZ8yYQZcuXSrFM2fOHBISElzX48aNY/jw4bz44ovExsYSERHBfffdR3m5c/j5yiuv5PDhw0yePBmTyfSbq6rWrFnD+PHjyc3Ndd0/Y8YMZs6c6dqR/pe6dOnCU089VSmWp59+mqioKIKDg5k0aRJlZWWu++12O7NmzaJ58+b4+fnRuXNnPvjgg7PGJHLBqyiFb1+CrW85r5tdfno47L07wVF29ve7gRIgNwvx9eLlkZ15eWQX3lh7gNEL1vPQu9tIiAhg9SP9CfTVPHS5uBUWFjJlyhQ2bdpEUlISZrOZm266CbvdTkFBAf379+fYsWOsXLmS7du389hjj2G3O8/9+eSTT7jpppsYMmQIW7duJSkpiZ49e7raHjduHJs2bWLlypWsW7cOh8PBkCFDXMnLuVq9ejXJycmsXr2aJUuWsHjxYhYvXgzA8uXLadKkCTNnziQ1NZXU1NSzttWnTx/mzJlDcHCw6/6pU6cyYcIEdu/ezQ8//OC6d+vWrezYscN1tiFAUlISu3fvZs2aNSxbtozly5fz9NNPu+pnzZrF0qVLmTdvHj/99BOTJ09mzJgxfP311zX6zCIXFLMZwp3/8OeyP8AtC+DWpc4kKDAaTManI/r2dbNgfx9+PlHI7W+sp9zmHO46WVTOn1f8yOCOMfxpcFsPRyhyfm655ZZK1wsXLiQqKopdu3bx/fffk5GRwQ8//EB4eDgArVq1ct3717/+lVGjRlVKADp37gzAvn37WLlyJd999x19+vQB4N///jfx8fGsWLGCW2+99ZxjDAsLY+7cuVgsFtq0acP1119PUlISEydOJDw8HIvFQlBQEDExMb/ZltVqJSQkBJPJVOn+wMBABg0axKJFi+jRowcAixYton///rRo0aLS+xcuXIi/vz/t27dn5syZPProozzzzDOUl5fz7LPP8r///Y/evXsD0KJFC9auXcs///lP+vfvf86fWeSC4nDAJYOh6WXgEwz/vhX6PATjPoPARuAw/jBU9QC52dHsIp7+70+u5OeXPtuZxski47v9ROrSvn37GD16NC1atCA4ONg1/JSSksK2bdvo2rWrK/n5tW3btjFgwIAq63bv3o2Xlxe9evVylUVERNC6dWt2795doxjbt29faZfk2NhY0tPTa9TGuZg4cSLLli2jpKSEsrIy3n77bSZMmFDpns6dO1faUbt3794UFBRw5MgR9u/fT1FREddccw2BgYGu19KlSysNK4pcfByw7S048DWsmwt3fAT2CtjzGayc7JGI1APkZoVlFfx0PK/a+m/3ZdCpSahxAYnUsaFDh9KsWTMWLFhAXFwcdrudDh06UFZWhp/f2fe4+q3632I2m/n1cYZVDY95e1eeYGkymVzDcHVp6NCh+Pj48NFHH2G1WikvL2fEiBHn/P6CggLAOTTYuHHjSnU+Pjo4WS5ioU2h71SwlzuHuwKjoNUgaHkV2O0Q2sTwkNQD5GYWs4mzzan0tyoHlYtXVlYWe/bs4cknn2TAgAG0bduWkydPuuo7derEtm3byM7OrvL9nTp1Iikpqcq6tm3bUlFRUWnC9KnntWvXDnCeg5WWllYpCdq2bVuNP4fVasVmO/dluNXd7+XlxZ133smiRYtYtGgRo0aNOiPJ2759O8XFxa7r9evXExgYSHx8PO3atcPHx4eUlBRatWpV6RUfb+wmcSJ1ruA4LBwEuSlgq4DyPPj3CEjbDkVV/x3hTkqA3CzY15v+iVFV1plMcHlLrQKTi1dYWBgRERHMnz+f/fv389VXXzFlyhRX/ejRo4mJiWH48OF89913HDhwgA8//JB169YBMH36dJYtW8b06dPZvXs3O3fu5PnnnwcgMTGRYcOGMXHiRNauXcv27dsZM2YMjRs3ZtiwYYBzBVdGRgazZ88mOTmZ1157jc8++6zGnyMhIYFvvvmGY8eOkZn523tzJSQkUFBQQFJSEpmZmRQVFbnqfv/73/PVV1+xatWqM4a/wHm47F133cWuXbv49NNPmT59Ovfffz9ms5mgoCCmTp3K5MmTWbJkCcnJyWzZsoVXX32VJUuW1PhziVww8tNg+T1w8iAsHeZcDr9sNJz4CT6Y4BwOM5gSIDdrFOzLn69vS1Tgmd3XfxzchlB/9QDJxctsNvPOO++wefNmOnTowOTJk3nhhRdc9VarlS+++IJGjRoxZMgQOnbsyHPPPeeaj3PllVfy/vvvs3LlSrp06cLVV1/Nxo0bXe9ftGgR3bp144YbbqB37944HA4+/fRT15BW27Zt+cc//sFrr71G586d2bhxI1OnTq3x55g5cyaHDh2iZcuWREVV/Q+WX+rTpw+TJk1i5MiRREVFMXv2bFddYmIiffr0oU2bNpXmL50yYMAAEhMTueKKKxg5ciQ33ngjM2bMcNU/88wzPPXUU8yaNYu2bdty3XXX8cknn9C8efMafy6RC0ZQDNzyhvPX0nx4+1bI+Bks3jDyX859ggxmcvx6AF3Iy8sjJCSE3NxcgoODz7u9jQeziAj04es96azdn0VUoA+jesZjsztoFORD04iAOohaxD1KSko4ePAgzZs3x9fX19PhXPAcDgeJiYn84Q9/qNQbBs5l/Tk5OaxYsaLOnqc/H7moHPkB3hx4+vrKP0Hv+8AnsE6ar8n3t7of3OxQZiGT393OsZxi7ruqJRMub05ecRn3/msTaXll/OuunkqAROqJjIwM3nnnHdLS0irt/SMiQPYBWHl/5bJ1cyFxIER3cp4YbyANgblZSbmNYznOCY+vrU5mzJsb+MPbW0nLcy5/33okx4PRiUhVBg8eXGkZ+i9fzz77bLXva9SoETNnzmT+/PmEhYUZGLHIBS4/FT646/Sw16BZ/z8clgdLh0Ox8ediqgfIzbwtZvx9zMy7vTuxIb5kF5bh523Bx9vM7FU/ExdyfsuARaTuvfHGG5VWav1SdXsaAWcsyf+1U7tPizQ4vqEw6K+wbBTc8iY07e1cAv/WcOg7xblRosGUALlZZJCVTx64gkXfHeTtDSlU/P/hp60aBfLq6K6E+Rt/AJyInN2v9+ARkfPk7QdxXeGBzWD2ds75adQW7v7auQt0cJzhIWkIzM18LGY+3nGcpesOu5IfgP3pBYxduJHicuO3/xYRETGctx8ERIFf6OmyoBiPJD+gBMjtjueW8Obag1XWZeSXsvdEvsERiYiIiBIgNyutsJNTVP3J1XvSlACJiIgYTQmQm/l4mQnxq36ezyXRdbP3gYiIiJw7JUBuFuLvxYTLE6qsiwy0ckl0kLEBiYiIiBIgd8vIL+OGTnGM6dUUi/n0qagtowJYOqEn+9I1BCZihCuvvJKHH37Y02GIyAVCy+DdzNti5sbXvuWft3fnzj4JZBeW4ettwdfbzIuf7+G6DrGeDlFERKTBUQLkZo0CfRjeuTFj3tx4Rp2ft4U/DmnrgahEPMtmd7DxYDbp+SU0CvKlZ/PwSj2kIiLupiEwNwvy82ZS/1b0al5599gAq4U37uxObPCZp8SL1Gerfkyl7/NfMXrBeh56ZxujF6yn7/NfserHVLc/226389hjjxEeHk5MTIzrFPZDhw5hMpnYtm2b696cnBxMJhNr1qwBYM2aNZhMJj7//HO6du2Kn58fV199Nenp6Xz22We0bduW4OBgbrvtNoqKik5/3lWr6Nu3L6GhoURERHDDDTeQnJzsqj/17OXLl3PVVVfh7+9P586dWbdundt/P0QMV5hV+bogwzNxoATIEPHh/rx4a2c++kMfZt7Yntdvv5T/PtCXS+ND8PfRTtDScKz6MZV7/7WF1NySSuVpuSXc+68tbk+ClixZQkBAABs2bGD27NnMnDmTL7/8skZtzJgxg7lz5/L9999z5MgRfve73zFnzhzefvttPvnkE7744gteffVV1/2FhYVMmTKFTZs2kZSUhNls5qabbsJur7wJ6p///GemTp3Ktm3buOSSSxg9ejQVFRV18rlFLgh5J8DsBYX/f+5XznEwWyDniEfC0RCYQeLD/YkP96drUx2QKA2Tze7g6f/uoqoTfxyACXj6v7u4pl2M24bDOnXqxPTp0wFITExk7ty5JCUlkZiYeM5t/OUvf+Hyyy8H4K677uKJJ54gOTmZFi1aADBixAhWr17N448/DsAtt9xS6f0LFy4kKiqKXbt20aFDB1f51KlTuf766wF4+umnad++Pfv376dNmza1/8AiF4r8DLAVQ0Gq81wwL19nF0zOEfALgdxUCDF2Tqx6gETEEBsPZp/R8/NLDiA1t4SNB7PdFkOnTp0qXcfGxpKenl7rNqKjo/H393clP6fKftnmvn37GD16NC1atCA4OJiEhAQAUlJSqm03Ntb5RVDT2EQuWBXFsPdzmN8ftv0bygrhxC54cwB8MQ3sZYaHpB4gETFEen71yU9t7qsNb+/KQ84mkwm73Y7Z7Py34C9Pcy8vr3oH91+2YTKZqm3zlKFDh9KsWTMWLFhAXFwcdrudDh06UFZW+S/8X7cLnDFMJnLRMgFHNkK/qfDNbDi+FQ6ugZ6TIGUtVNk37F5KgETEEI2CfOv0vroUFRUFQGpqKl27dgWoNCG6trKystizZw8LFiygX79+AKxdu/a82xW56PiHw4BpkH8c2g+HeX3hpn9CeEvoOREsxi8IUgIkIobo2Tyc2BBf0nJLqvy3ngmICXEuiTean58fl112Gc899xzNmzcnPT2dJ5988rzbDQsLIyIigvnz5xMbG0tKSgp//OMf6yBikYtMYTb89BH8bzr0fgAm74LU7bBoELQcCENeMDwkzQESEUNYzCamD20HOJOdXzp1PX1oO4/tB7Rw4UIqKiro1q0bDz/8MH/5y1/Ou02z2cw777zD5s2b6dChA5MnT+aFF4z/i17E8+yQudf547pX4aN74L0xYLc5e4U8MARmcvxy0FsAyMvLIyQkhNzcXIKDgz0djohHlZSUcPDgQZo3b46v7/kPT636MZWn/7ur0oTo2BBfpg9tp53Ra6Gu/3xE3CL3GJQXw9qXnJOgT4npCLcuAS//OlkFVpPvbw2BGeREXjF5xRWk55cS4GMhzN9Ks4gAT4clYrjrOsRyTbsY7QQt0pCENIa8Y9BqYOUEKLYrWKzga/zB4EqADHA0u4g5SftYvuUo9v/vb4sP9+P127vRoXGIZ4MT8QCL2UTvlhGeDkNEjJKfBse3wfLfO68t3mArh61LwTcYet0DPoGGhqQ5QG5WXFbBe5uO8MHm08kPwJHsYsYu3MjBzALPBSciImIEuw3Wv+78NaYjTP4Juo5x1v34AWfODHQ/9QC5WWpuCYu/P1RlXXZhGT+n5tM80tisV0RExFAhjWHYXPju79DnQQiMhv6Pg38EdB4NofGGh6QeIDcrs9nJK6n+PJ996eoBEhGRBiAsAa58AsKbO69Dm0KvSdCorUfCUQLkZlaLmfAAa7X1bWKMn/glIiLiEYGNKl8Hx3kmDjQE5nZhflZ+37c5X+0+wZM3tCPQ1wub3cHqn0+w6PvDJDbS8JeIiIjRlAC5WUZhKTd2jqFn83Ce/3wPWw6fJCLAytg+Cbw/qTf70vJoHqUkSERExEhKgNws0MfCwcwi7nhzg2sV2PHcEp777GfWH8jiqRs8M/YpIiLSkGkOkJuV2xw88/GuSkvgT1mzJ4Pc4uonSItI3bnyyit5+OGHPR2GSMNls0FBhnNH6FPy05xlHqAEyM1Kym38nJZfbf33+zMNjEZEamvx4sWEhoZ6OgyRi5PNBid2wj/7Qeo2KC+FrP2weAgc/NojSZASIDezmM1n3eI/2NfbwGhELgAVpXDwGzh1DKHD4byuKPVsXCLiPkUZsPJ+yE+Ft26Gnz+Gf98KWcnOcuyGh6QEyM2CfCwMaNOoyjqzCR0HIA1LRSksuw2WDIVVT4DdDqv+6Lxedpvbk6CKigruv/9+QkJCiIyM5KmnnuLUedAnT55k7NixhIWF4e/vz+DBg9m3bx8Aa9asYfz48eTm5mIymTCZTMyYMcOtsYrUK0ExMGKhc++f8iL4cAJkHwAvH7jtPfA2/mxMJUBuVlxh4/Hr2hAXcuYpzc8M64CX/gSkoTiV/Bz4ynm94XVnd/iGec7rA1+5PQlasmQJXl5ebNy4kb///e+89NJLvPHGGwCMGzeOTZs2sXLlStatW4fD4WDIkCGUl5fTp08f5syZQ3BwMKmpqaSmpjJ16lS3xSlSLwXGwohFlcv6PgLRHQ0/BwwukATotddeIyEhAV9fX3r16sXGjRvPev/7779PmzZt8PX1pWPHjnz66aeV6h0OB9OmTSM2NhY/Pz8GDhzo+pec0fy9LGw/epK3ft+LF0Z04sbOcUzs25xPHuhL41Bf7B44/0TEI45sgOT/geMXXd0nfjz9s8PurD9y9v//z0d8fDwvv/wyrVu35vbbb+eBBx7g5ZdfZt++faxcuZI33niDfv360blzZ/79739z7NgxVqxYgdVqJSQkBJPJRExMDDExMQQGavsKkXNWmAUFabB8YuXy7+ZAxs9QaPx8WI8nQO+++y5Tpkxh+vTpbNmyhc6dOzNo0CDS09OrvP/7779n9OjR3HXXXWzdupXhw4czfPhwfvzx9F+ks2fP5pVXXmHevHls2LCBgIAABg0aRElJiVEfyyW/zMaMlbsY8Lev2ZJykqGdY2nZKJCR/1zHuMWb2JOWZ3hMIh6R0M+57f3Z9LoXEvq6LYTLLrsMk+n0Pzp69+7Nvn372LVrF15eXvTq1ctVFxERQevWrdm9e7fb4hFpMOxl8MEvhr2GvXZ6OOzftzgPSTWYxxOgl156iYkTJzJ+/HjatWvHvHnz8Pf3Z+HChVXe//e//53rrruORx99lLZt2/LMM89w6aWXMnfuXMDZ+zNnzhyefPJJhg0bRqdOnVi6dCnHjx9nxYoVBn4yp/JfnAW2bOMRJi7dzB+X76SgzPmHvS+90PCYRDzCZIJBsyC6Q9X10R1g0LPO+0SkfvEKgBvmQEAk3P4B+IU5fw1tBoNfAFuZ4SF5NAEqKytj8+bNDBw40FVmNpsZOHAg69atq/I969atq3Q/wKBBg1z3Hzx4kLS0tEr3hISE0KtXr2rbLC0tJS8vr9KrrvzWWWDtYnUWmDQQDgd8/kTlYa9fOvEjfP6n06vD3GDDhg2VrtevX09iYiLt2rWjoqKiUn1WVhZ79uyhXbt2AFitVmw24/+VKlIv+AVDRCv4w3o4vg3euQ0+exzu+gKa9294p8FnZmZis9mIjo6uVB4dHU1aWlqV70lLSzvr/ad+rUmbs2bNIiQkxPWKj6+7P4ggXy/uvqJFlXUxwb601DEY0lAc+vb0hOfqbHgdDq11WwgpKSlMmTKFPXv2sGzZMl599VUeeughEhMTGTZsGBMnTmTt2rVs376dMWPG0LhxY4YNGwZAQkICBQUFJCUlkZmZSVFRkdviFKl38lKhOBveuR2+fMpZdmC1cyFEaR4UNMA5QBeCJ554gtzcXNfryJEjdda2l8XMte2imXRFC3x+seSrQ+NglkzoiZ/VUmfPErmgxfeClgPB9Iu/dn45HGYyO+vje7othLFjx1JcXEzPnj257777eOihh7j77rsBWLRoEd26deOGG26gd+/eOBwOPv30U7y9nXt19enTh0mTJjFy5EiioqKYPXu22+IUqXccNvjPfc7FECYT9L4ffEOgIB3eGga24t9uo4559CywyMhILBYLJ06cqFR+4sQJYmJiqnxPTEzMWe8/9euJEyeIjY2tdE+XLl2qbNPHxwcfH5/afoyzCvO3UlBSzlVtorilWxNyi8vx9bZgMUO5rYIAa/XDYyL1ipcPjH7budQ9+X/OCc+DnnUOi22YBy2udtZ7uef/xTVr1rh+fv3118+oDwsLY+nSpWdt4/XXX6/yvSLyW8zQ5wE4uhGG/QNaXg0dR8DSYdD5dqo8L8r9EXmO1WqlW7duJCUlucrsdjtJSUn07t27yvf07t270v0AX375pev+5s2bExMTU+mevLw8NmzYUG2b7hYfHkBsiB+FZTb2pOVzPKcYq5eFJuF+BPu55y97kQvSqSTozo/hullgNsN1zzmv3Zj8iIiHhcRBTBd4aAc0vcw5GTokASathe4TIKyp4SF5/DT4KVOmcOedd9K9e3d69uzJnDlzKCwsZPz48YCzy7px48bMmjULgIceeoj+/fvzt7/9jeuvv5533nmHTZs2MX/+fABMJhMPP/wwf/nLX0hMTKR58+Y89dRTxMXFMXz4cE99TJpGBBAf7qBtdBBmiwlvi0YfpYHy8oHm/U5fm0yVr0WkfgqJdZ4JZvn/qR8BYWALPn1tMI8nQCNHjiQjI4Np06aRlpZGly5dWLVqlWsSc0pKCmbz6WShT58+vP322zz55JP86U9/IjExkRUrVtChw+m5BI899hiFhYXcfffd5OTk0LdvX1atWoWv75m7MRvJZDLhozk/IiLSUP062fFQ8gNgcjjcuOb0IpWXl0dISAi5ubkEBwd7OhwRjyopKeHgwYM0b97c4/+IkDPpz0fktJp8f2scRkTOif6tdGHSn4tI7SgBEpGzOrUMXPveXJjKypw76Fo8OJQgcjHy+BwgEbmwWSwWQkNDXefz+fv7VzpPSzzHbreTkZGBv78/Xl7661ykJvR/jIj8plP7a1V3SLF4jtlspmnTpkpKRWpICZCI/CaTyURsbCyNGjWivLzc0+HIL1it1korZUXk3CgBEpFzZrFYNNdEROoF/bNBREREGhwlQCIiItLgKAESERGRBkdzgKpwamOxvLw8D0ciIiIi5+rU9/a5bBCqBKgK+fn5AMTHx3s4EhEREamp/Px8QkJCznqPzgKrgt1u5/jx4wQFBdX53hp5eXnEx8dz5MgRnTMmIiINkru+Cx0OB/n5+cTFxf3m9hDqAaqC2WymSZMmbn1GcHCwEiAREWnQ3PFd+Fs9P6doErSIiIg0OEqAREREpMFRAmQwHx8fpk+fjo+Pj6dDERER8YgL4btQk6BFRESkwVEPkIiIiDQ4SoBERESkwVECJCIiIg2OEqBqXHnllTz88MOeDkNEROSicTF9dyoBEhERkQvC4sWLCQ0NNeRZSoBERESkwVECdBZ2u53HHnuM8PBwYmJimDFjBgCHDh3CZDKxbds21705OTmYTCbWrFkDwJo1azCZTHz++ed07doVPz8/rr76atLT0/nss89o27YtwcHB3HbbbRQVFbnaWbVqFX379iU0NJSIiAhuuOEGkpOTXfWnnr18+XKuuuoq/P396dy5M+vWrTPit0REROSsKioquP/++wkJCSEyMpKnnnrKdTr7yZMnGTt2LGFhYfj7+zN48GD27dsHOL83x48fT25uLiaTCZPJ5PredQclQGexZMkSAgIC2LBhA7Nnz2bmzJl8+eWXNWpjxowZzJ07l++//54jR47wu9/9jjlz5vD222/zySef8MUXX/Dqq6+67i8sLGTKlCls2rSJpKQkzGYzN910E3a7vVK7f/7zn5k6dSrbtm3jkksuYfTo0VRUVNTJ5xYREamtJUuW4OXlxcaNG/n73//OSy+9xBtvvAHAuHHj2LRpEytXrmTdunU4HA6GDBlCeXk5ffr0Yc6cOQQHB5OamkpqaipTp051X6AOqVL//v0dffv2rVTWo0cPx+OPP+44ePCgA3Bs3brVVXfy5EkH4Fi9erXD4XA4Vq9e7QAc//vf/1z3zJo1ywE4kpOTXWX33HOPY9CgQdXGkZGR4QAcO3fudDgcDtez33jjDdc9P/30kwNw7N69+3w+soiIyHnp37+/o23btg673e4qe/zxxx1t27Z17N271wE4vvvuO1ddZmamw8/Pz/Hee+85HA6HY9GiRY6QkBBDYlUP0Fl06tSp0nVsbCzp6em1biM6Ohp/f39atGhRqeyXbe7bt4/Ro0fTokULgoODSUhIACAlJaXadmNjYwFqHJuIiEhdu+yyyzCZTK7r3r17s2/fPnbt2oWXlxe9evVy1UVERNC6dWt2795teJxehj/xIuLt7V3p2mQyYbfbMZudeaPjF6eIlJeX/2YbJpOp2jZPGTp0KM2aNWPBggXExcVht9vp0KEDZWVlZ20XOGOYTERERKqmHqBaiIqKAiA1NdVV9ssJ0bWVlZXFnj17ePLJJxkwYABt27bl5MmT592uiIiIUTZs2FDpev369SQmJtKuXTsqKioq1Z/63mvXrh0AVqsVm81mSJxKgGrBz8+Pyy67jOeee47du3fz9ddf8+STT553u2FhYURERDB//nz279/PV199xZQpU+ogYhEREWOkpKQwZcoU9uzZw7Jly3j11Vd56KGHSExMZNiwYUycOJG1a9eyfft2xowZQ+PGjRk2bBgACQkJFBQUkJSURGZmZqVV0nVNCVAtLVy4kIqKCrp168bDDz/MX/7yl/Nu02w2884777B582Y6dOjA5MmTeeGFF+ogWhEREWOMHTuW4uJievbsyX333cdDDz3E3XffDcCiRYvo1q0bN9xwA71798bhcPDpp5+6pnX06dOHSZMmMXLkSKKiopg9e7bb4jQ5fjmRRURERKQBUA+QiIiINDhKgERERKTBUQIkIiIiDY4SIBEREWlwlACJiIhIg6MESERERBocJUAiIiLS4CgBEhERkQZHCZCIiIg0OEqARKTBS0hIYM6cOed8/+LFiwkNDXVbPCLifkqAREREpMFRAiQi52XVqlX07duX0NBQIiIiuOGGG0hOTnbVHz16lNGjRxMeHk5AQADdu3dnw4YNrvr//ve/9OjRA19fXyIjI7nppptcdSdPnmTs2LGEhYXh7+/P4MGD2bdvn6t+xowZdOnSpVI8c+bMISEhwXU9btw4hg8fzosvvkhsbCwRERHcd999lJeXA3DllVdy+PBhJk+ejMlkwmQynfXzrlmzhvHjx5Obm+u6f8aMGcycOZMOHTqccX+XLl146qmnKsXy9NNPExUVRXBwMJMmTaKsrMx1v91uZ9asWTRv3hw/Pz86d+7MBx98cNaYRKTmlACJyHkpLCxkypQpbNq0iaSkJMxmMzfddBN2u52CggL69+/PsWPHWLlyJdu3b+exxx7DbrcD8Mknn3DTTTcxZMgQtm7dSlJSEj179nS1PW7cODZt2sTKlStZt24dDoeDIUOGuJKXc7V69WqSk5NZvXo1S5YsYfHixSxevBiA5cuX06RJE2bOnElqaiqpqalnbatPnz7MmTOH4OBg1/1Tp05lwoQJ7N69mx9++MF179atW9mxYwfjx493lSUlJbF7927WrFnDsmXLWL58OU8//bSrftasWSxdupR58+bx008/MXnyZMaMGcPXX39do88sIr/BISJShzIyMhyAY+fOnY5//vOfjqCgIEdWVlaV9/bu3dtx++23V1m3d+9eB+D47rvvXGWZmZkOPz8/x3vvvedwOByO6dOnOzp37lzpfS+//LKjWbNmrus777zT0axZM0dFRYWr7NZbb3WMHDnSdd2sWTPHyy+/fM6fcdGiRY6QkJAzygcPHuy49957XdcPPPCA48orr6wUS3h4uKOwsNBV9vrrrzsCAwMdNpvNUVJS4vD393d8//33ldq96667HKNHjz7n+ETkt6kHSETOy759+xg9ejQtWrQgODjYNfyUkpLCtm3b6Nq1K+Hh4VW+d9u2bQwYMKDKut27d+Pl5UWvXr1cZREREbRu3Zrdu3fXKMb27dtjsVhc17GxsaSnp9eojXMxceJEli1bRklJCWVlZbz99ttMmDCh0j2dO3fG39/fdd27d28KCgo4cuQI+/fvp6ioiGuuuYbAwEDXa+nSpZWGFUXk/Hl5OgARubgNHTqUZs2asWDBAuLi4rDb7XTo0IGysjL8/PzO+t7fqv8tZrMZh8NRqayq4TFvb+9K1yaTyTUMV5eGDh2Kj48PH330EVarlfLyckaMGHHO7y8oKACcQ4ONGzeuVOfj41OnsYo0dEqARKTWsrKy2LNnDwsWLKBfv34ArF271lXfqVMn3njjDbKzs6vsBerUqRNJSUmV5sic0rZtWyoqKtiwYQN9+vSp9Lx27doBEBUVRVpaGg6HwzV5edu2bTX+HFarFZvNdt73e3l5ceedd7Jo0SKsViujRo06I8nbvn07xcXFrvL169cTGBhIfHw84eHh+Pj4kJKSQv/+/Wv8OUTk3GkITERqLSwsjIiICObPn8/+/fv56quvmDJliqt+9OjRxMTEMHz4cL777jsOHDjAhx9+yLp16wCYPn06y5YtY/r06ezevZudO3fy/PPPA5CYmMiwYcOYOHEia9euZfv27YwZM4bGjRszbNgwwLmCKyMjg9mzZ5OcnMxrr73GZ599VuPPkZCQwDfffMOxY8fIzMw8p/sLCgpISkoiMzOToqIiV93vf/97vvrqK1atWnXG8BdAWVkZd911F7t27eLTTz9l+vTp3H///ZjNZoKCgpg6dSqTJ09myZIlJCcns2XLFl599VWWLFlS488lImfh6UlIInJx+/LLLx1t27Z1+Pj4ODp16uRYs2aNA3B89NFHDofD4Th06JDjlltucQQHBzv8/f0d3bt3d2zYsMH1/g8//NDRpUsXh9VqdURGRjpuvvlmV112drbjjjvucISEhDj8/PwcgwYNcuzdu7fS819//XVHfHy8IyAgwDF27FjHX//61zMmQQ8bNqzSex566CFH//79Xdfr1q1zdOrUyeHj4+M4178WJ02a5IiIiHAAjunTp1eq69evn6N9+/ZnvOdULNOmTXNEREQ4AgMDHRMnTnSUlJS47rHb7Y45c+Y4Wrdu7fD29nZERUU5Bg0a5Pj666/PKS4ROTcmh+NXA+giIlJrDoeDxMRE/vCHP1TqDQPnsv6cnBxWrFjhmeBExEVzgERE6khGRgbvvPMOaWlpVc5rEpELh+YAiYj8yuDBgystQ//l69lnn632fY0aNWLmzJnMnz+fsLAwAyMWkZrSEJiIyK8cO3aM4uLiKuvCw8Or3ddIRC4eSoBERESkwdEQmIiIiDQ4SoBERESkwVECJCIiIg2OEiARERFpcJQAiYiISIOjBEhEREQaHCVAIiIi0uD8H/5OyrXtqlmwAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data = df_mean_var_w_label, x = \"account_type\", y = \"variance_sentiment\", hue = \"account_type\", style = \"account_type\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d6824edc-b4ce-4718-b331-34ad71f618d2",
   "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>userid</th>\n",
       "      <th>account_type</th>\n",
       "      <th>clean_tweet</th>\n",
       "      <th>vader_sentiment</th>\n",
       "      <th>vader_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can put the \"River\" in \"Screwdrivers\" https...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can't extract the \"Pong\" from \"Sponged\" ht...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>Put the \"Nope\" back in \"Inoperative\" https://t...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can't spell \"Artworks\" without \"Two\" https:...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can be the \"She\" in \"Astonished\" https://t...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               userid account_type  \\\n",
       "0  787405734442958848          bot   \n",
       "1  787405734442958848          bot   \n",
       "2  787405734442958848          bot   \n",
       "3  787405734442958848          bot   \n",
       "4  787405734442958848          bot   \n",
       "\n",
       "                                         clean_tweet  vader_sentiment  \\\n",
       "0  We can put the \"River\" in \"Screwdrivers\" https...              0.0   \n",
       "1  You can't extract the \"Pong\" from \"Sponged\" ht...              0.0   \n",
       "2  Put the \"Nope\" back in \"Inoperative\" https://t...              0.0   \n",
       "3  We can't spell \"Artworks\" without \"Two\" https:...              0.0   \n",
       "4  You can be the \"She\" in \"Astonished\" https://t...              0.0   \n",
       "\n",
       "  vader_label  \n",
       "0     neutral  \n",
       "1     neutral  \n",
       "2     neutral  \n",
       "3     neutral  \n",
       "4     neutral  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a86b48cf-f9cb-44fd-b919-85992a72b7e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_url_flag(in_tweet):\n",
    "    m = re.search(r\"http[a-zA-Z0-9/\\-.:%]+\", in_tweet)\n",
    "    rtn = False\n",
    "    if m:\n",
    "        rtn = True\n",
    "    return rtn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "15699282-abee-4392-9f1d-9d4f67c1a8a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1.4 s, sys: 7.89 ms, total: 1.41 s\n",
      "Wall time: 1.46 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "df['url_flag'] = df['clean_tweet'].apply(get_url_flag)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4d470cc5-743a-49f6-b715-3ecbc21701de",
   "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>userid</th>\n",
       "      <th>account_type</th>\n",
       "      <th>clean_tweet</th>\n",
       "      <th>vader_sentiment</th>\n",
       "      <th>vader_label</th>\n",
       "      <th>url_flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can put the \"River\" in \"Screwdrivers\" https...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can't extract the \"Pong\" from \"Sponged\" ht...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>Put the \"Nope\" back in \"Inoperative\" https://t...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can't spell \"Artworks\" without \"Two\" https:...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can be the \"She\" in \"Astonished\" https://t...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488632</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>A Tribute – Joan Rivers on Fitness http://t.co...</td>\n",
       "      <td>0.2732</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488633</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Live in Orlando area? Want a personal trainer?...</td>\n",
       "      <td>0.1680</td>\n",
       "      <td>positive</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488634</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>If you're trying to get fit  keep working. You...</td>\n",
       "      <td>0.8608</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488635</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Ready for summer? Cheap  cool sunglasses - loo...</td>\n",
       "      <td>0.7430</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488636</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Hey all. Stayn' healthy is important especiall...</td>\n",
       "      <td>0.5423</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>488637 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    userid account_type  \\\n",
       "0       787405734442958848          bot   \n",
       "1       787405734442958848          bot   \n",
       "2       787405734442958848          bot   \n",
       "3       787405734442958848          bot   \n",
       "4       787405734442958848          bot   \n",
       "...                    ...          ...   \n",
       "488632          3077047801          bot   \n",
       "488633          3077047801          bot   \n",
       "488634          3077047801          bot   \n",
       "488635          3077047801          bot   \n",
       "488636          3077047801          bot   \n",
       "\n",
       "                                              clean_tweet  vader_sentiment  \\\n",
       "0       We can put the \"River\" in \"Screwdrivers\" https...           0.0000   \n",
       "1       You can't extract the \"Pong\" from \"Sponged\" ht...           0.0000   \n",
       "2       Put the \"Nope\" back in \"Inoperative\" https://t...           0.0000   \n",
       "3       We can't spell \"Artworks\" without \"Two\" https:...           0.0000   \n",
       "4       You can be the \"She\" in \"Astonished\" https://t...           0.0000   \n",
       "...                                                   ...              ...   \n",
       "488632  A Tribute – Joan Rivers on Fitness http://t.co...           0.2732   \n",
       "488633  Live in Orlando area? Want a personal trainer?...           0.1680   \n",
       "488634  If you're trying to get fit  keep working. You...           0.8608   \n",
       "488635  Ready for summer? Cheap  cool sunglasses - loo...           0.7430   \n",
       "488636  Hey all. Stayn' healthy is important especiall...           0.5423   \n",
       "\n",
       "       vader_label  url_flag  \n",
       "0          neutral      True  \n",
       "1          neutral      True  \n",
       "2          neutral      True  \n",
       "3          neutral      True  \n",
       "4          neutral      True  \n",
       "...            ...       ...  \n",
       "488632    positive      True  \n",
       "488633    positive     False  \n",
       "488634    positive      True  \n",
       "488635    positive      True  \n",
       "488636    positive      True  \n",
       "\n",
       "[488637 rows x 6 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "22d3b489-b23e-41fd-b059-258ab3871bed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'A Tribute – Joan Rivers on Fitness http://t.co/eZQVBugIC2 via @fittipdaily'"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[488632]['clean_tweet']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "715fa405-dc45-4629-a488-fd77966cdaaf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Live in Orlando area? Want a personal trainer? This woman knows her stuff &amp; priced right: @mytrainertracey  #personaltrainer #healthyliving'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[488633]['clean_tweet']"
   ]
  },
  {
   "cell_type": "raw",
   "id": "672e0884-8cce-49ae-935e-eceda50bfb78",
   "metadata": {},
   "source": [
    "%%time\n",
    "\n",
    "# Takes infinite time\n",
    "\n",
    "sns.scatterplot(data = df, x = \"account_type\", y = \"url_flag\", hue = \"account_type\", style = \"account_type\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "745068aa-947f-48c5-ae42-cec761509c55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "account_type  url_flag\n",
       "bot           True         68899\n",
       "              False        47947\n",
       "human         False       202148\n",
       "              True        169643\n",
       "Name: url_flag, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['account_type'])['url_flag'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "367ff1c7-ee2e-4c2b-8dba-708399244a80",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5896564709104292\n",
      "0.4562859240810025\n"
     ]
    }
   ],
   "source": [
    "print(68899 / (68899+47947))\n",
    "print(169643 / (169643 + 202148))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "4becbe0b-010f-449c-8f9e-feb2658808c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['len'] = df['clean_tweet'].apply(len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "00bf761b-e943-4070-8cb7-7b81bab5a30e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "account_type\n",
       "bot      111.761455\n",
       "human    109.129188\n",
       "Name: len, dtype: float64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['account_type'])['len'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "4b70873d-8103-4f8c-b862-65bc130a360b",
   "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>userid</th>\n",
       "      <th>account_type</th>\n",
       "      <th>clean_tweet</th>\n",
       "      <th>vader_sentiment</th>\n",
       "      <th>vader_label</th>\n",
       "      <th>url_flag</th>\n",
       "      <th>len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can put the \"River\" in \"Screwdrivers\" https...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can't extract the \"Pong\" from \"Sponged\" ht...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "      <td>67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>Put the \"Nope\" back in \"Inoperative\" https://t...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>We can't spell \"Artworks\" without \"Two\" https:...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>787405734442958848</td>\n",
       "      <td>bot</td>\n",
       "      <td>You can be the \"She\" in \"Astonished\" https://t...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>True</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488632</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>A Tribute – Joan Rivers on Fitness http://t.co...</td>\n",
       "      <td>0.2732</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488633</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Live in Orlando area? Want a personal trainer?...</td>\n",
       "      <td>0.1680</td>\n",
       "      <td>positive</td>\n",
       "      <td>False</td>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488634</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>If you're trying to get fit  keep working. You...</td>\n",
       "      <td>0.8608</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488635</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Ready for summer? Cheap  cool sunglasses - loo...</td>\n",
       "      <td>0.7430</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>488636</th>\n",
       "      <td>3077047801</td>\n",
       "      <td>bot</td>\n",
       "      <td>Hey all. Stayn' healthy is important especiall...</td>\n",
       "      <td>0.5423</td>\n",
       "      <td>positive</td>\n",
       "      <td>True</td>\n",
       "      <td>135</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>488637 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    userid account_type  \\\n",
       "0       787405734442958848          bot   \n",
       "1       787405734442958848          bot   \n",
       "2       787405734442958848          bot   \n",
       "3       787405734442958848          bot   \n",
       "4       787405734442958848          bot   \n",
       "...                    ...          ...   \n",
       "488632          3077047801          bot   \n",
       "488633          3077047801          bot   \n",
       "488634          3077047801          bot   \n",
       "488635          3077047801          bot   \n",
       "488636          3077047801          bot   \n",
       "\n",
       "                                              clean_tweet  vader_sentiment  \\\n",
       "0       We can put the \"River\" in \"Screwdrivers\" https...           0.0000   \n",
       "1       You can't extract the \"Pong\" from \"Sponged\" ht...           0.0000   \n",
       "2       Put the \"Nope\" back in \"Inoperative\" https://t...           0.0000   \n",
       "3       We can't spell \"Artworks\" without \"Two\" https:...           0.0000   \n",
       "4       You can be the \"She\" in \"Astonished\" https://t...           0.0000   \n",
       "...                                                   ...              ...   \n",
       "488632  A Tribute – Joan Rivers on Fitness http://t.co...           0.2732   \n",
       "488633  Live in Orlando area? Want a personal trainer?...           0.1680   \n",
       "488634  If you're trying to get fit  keep working. You...           0.8608   \n",
       "488635  Ready for summer? Cheap  cool sunglasses - loo...           0.7430   \n",
       "488636  Hey all. Stayn' healthy is important especiall...           0.5423   \n",
       "\n",
       "       vader_label  url_flag  len  \n",
       "0          neutral      True   64  \n",
       "1          neutral      True   67  \n",
       "2          neutral      True   60  \n",
       "3          neutral      True   63  \n",
       "4          neutral      True   60  \n",
       "...            ...       ...  ...  \n",
       "488632    positive      True   74  \n",
       "488633    positive     False  143  \n",
       "488634    positive      True  138  \n",
       "488635    positive      True  133  \n",
       "488636    positive      True  135  \n",
       "\n",
       "[488637 rows x 7 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dcdccebd-4ed0-435d-b888-e11f3eedeec3",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "c5f10c2e-437c-4ef3-8e68-dfc576a36d13",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = RandomForestClassifier(random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ffe95aaa-7116-4090-9964-dc616b2eb70b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[['vader_sentiment', 'url_flag', 'len']]\n",
    "y = df['account_type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "cd67098b-eeca-4ab3-9f33-28aa7cc17cd2",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f838ecfe-4522-427f-a41d-ee67481b1b4e",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "4293be80-317a-4131-bce1-17ee63176426",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "569197c6-f69f-436c-bc6d-d16a0cf68ee8",
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = ['bot', 'human']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "47f8a948-f997-499d-aa36-a4fc93ae062f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         bot       0.54      0.24      0.33     38217\n",
      "       human       0.80      0.93      0.86    123034\n",
      "\n",
      "    accuracy                           0.77    161251\n",
      "   macro avg       0.67      0.59      0.60    161251\n",
      "weighted avg       0.74      0.77      0.74    161251\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred = pred, labels = labels))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "041aed1d-2d82-4f86-9541-0beee16426e5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "024dd6fa-545b-4d8e-9579-9e6a018fd866",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "rasa_py38",
   "language": "python",
   "name": "rasa_py38"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}