05_Exploration-and-Visualization_limited_merged_negative-binary-CLASS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "41393559-452c-4856-b494-3d8d3bf42225",
   "metadata": {},
   "source": [
    "# Exploration and Visualization\n",
    "**Author**: Andrea Cass"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ffdd371-590c-4d74-b86a-d8bab76f00c1",
   "metadata": {},
   "source": [
    "## 1. About this notebook\n",
    "The purpose of this notebook is to explore and visualize some of the prepared data,\n",
    "\n",
    "> *04_Prepared-data_limited_merged.csv*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b5ab363-17cd-44ea-9a30-942e04500702",
   "metadata": {},
   "source": [
    "## 2. Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e7b28001-edd1-4828-8c90-86389dc2eade",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import csv\n",
    "import os\n",
    "from pathlib import Path\n",
    "from datetime import datetime"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5566d29c-54b0-4523-a52b-c748ba365a89",
   "metadata": {},
   "source": [
    "## 3. Working directory & file paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "859f4051-f04c-4cfb-a109-66f5ccf6500f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/Users/andycass/Jupyterlab_main-folder/THESIS/limited collection/revised/negative binary class'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find current working directory\n",
    "\n",
    "os.getcwd()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e2b249c2-2d6b-4455-b521-d15a9026c8a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# changing current working directory\n",
    "\n",
    "os.chdir('/Users/andycass/Desktop/Thesis_data-and-code')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5c4ad9e1-a4cf-44c8-9040-059cfc275578",
   "metadata": {},
   "outputs": [],
   "source": [
    "# naming the current working directory\n",
    "\n",
    "cwd = Path.cwd()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "894913dc-3dd4-42ad-80c5-ae41eb3ea766",
   "metadata": {},
   "outputs": [],
   "source": [
    "# naming the CASS_thesis folder\n",
    "\n",
    "CASS_thesis = cwd / 'CASS_thesis'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "71853d45-591d-46ee-bd00-b9cef4fddc80",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PosixPath('/Users/andycass/Desktop/Thesis_data-and-code/CASS_thesis')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# double-checking the CASS_thesis location\n",
    "\n",
    "CASS_thesis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5af243fa-6e29-4e76-923d-c2ee2065ef29",
   "metadata": {},
   "source": [
    "## 4. Exploring with categorical CLASS (Negative, Neutral, Positive)\n",
    "In this section, the dataset will be explored with CLASS as it currently is, a 3-part categorical variable inidcating whether a tweet is Negative, Neutral, or Positive.\n",
    "\n",
    "### 4.1. Exploring CLASS\n",
    "#### 4.1.1. Exploring CLASS for entire dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "54dbf7fb-fa35-4936-a15b-5d1e69303c59",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(CASS_thesis / '04_Prepared-data_limited_merged.csv', index_col=[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f5a1a2d3-3b80-40ba-a0be-c4934f84d0f1",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>Language</th>\n",
       "      <th>date</th>\n",
       "      <th>inflow</th>\n",
       "      <th>new text</th>\n",
       "      <th>scores</th>\n",
       "      <th>Negative</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>['0.0679844', '0.88380396', '0.0482117', '']</td>\n",
       "      <td>0.067984</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>['0.3660313', '0.5833739', '0.05059479']</td>\n",
       "      <td>0.366031</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>['0.30297568', '0.4430802', '0.25394407']</td>\n",
       "      <td>0.302976</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>['0.02260253', '0.07604674', '0.9013506', '']</td>\n",
       "      <td>0.022603</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>['0.6982864', '0.27545568', '0.02625783']</td>\n",
       "      <td>0.698286</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>66261</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-28 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>For day 1 of week 2, @user discusses the Socia...</td>\n",
       "      <td>['0.03550766', '0.82208896', '0.14240335']</td>\n",
       "      <td>0.035508</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66262</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-27 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>@user Europe is racist, but in a different way...</td>\n",
       "      <td>['0.8931986', '0.09380516', '0.01299612']</td>\n",
       "      <td>0.893199</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66263</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-27 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>['0.15655968', '0.5201203', '0.32332003']</td>\n",
       "      <td>0.156560</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66264</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-26 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>@user I must have missed the great flight of t...</td>\n",
       "      <td>['0.91320664', '0.06375945', '0.02303378']</td>\n",
       "      <td>0.913207</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66265</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-25 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>['0.92373', '0.06414817', '0.01212183']</td>\n",
       "      <td>0.923730</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "66261  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "66262  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "66263  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "66264  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "66265  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "66261  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "66262  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "66263  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "66264  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "66265  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "66261                           0.0                         1.0   \n",
       "66262                           0.0                         1.0   \n",
       "66263                           0.0                         1.0   \n",
       "66264                           0.0                         0.0   \n",
       "66265                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "66261                        2.0                         0.0   \n",
       "66262                        6.0                         0.0   \n",
       "66263                       17.0                         0.0   \n",
       "66264                        1.0                         0.0   \n",
       "66265                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ... Language  \\\n",
       "0                                  0.0  ...   German   \n",
       "1                                  0.0  ...   German   \n",
       "2                                  0.0  ...   German   \n",
       "3                                  0.0  ...   German   \n",
       "4                                  0.0  ...   German   \n",
       "...                                ...  ...      ...   \n",
       "66261                              0.0  ...  English   \n",
       "66262                              0.0  ...  English   \n",
       "66263                              0.0  ...  English   \n",
       "66264                              0.0  ...  English   \n",
       "66265                              0.0  ...  English   \n",
       "\n",
       "                            date      inflow  \\\n",
       "0      2016-04-20 00:00:00+00:00     Syrians   \n",
       "1      2016-04-20 00:00:00+00:00     Syrians   \n",
       "2      2016-04-20 00:00:00+00:00     Syrians   \n",
       "3      2016-04-20 00:00:00+00:00     Syrians   \n",
       "4      2016-04-20 00:00:00+00:00     Syrians   \n",
       "...                          ...         ...   \n",
       "66261  2021-06-28 00:00:00+00:00  Ukrainians   \n",
       "66262  2021-06-27 00:00:00+00:00  Ukrainians   \n",
       "66263  2021-06-27 00:00:00+00:00  Ukrainians   \n",
       "66264  2021-06-26 00:00:00+00:00  Ukrainians   \n",
       "66265  2021-06-25 00:00:00+00:00  Ukrainians   \n",
       "\n",
       "                                                new text  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...   \n",
       "...                                                  ...   \n",
       "66261  For day 1 of week 2, @user discusses the Socia...   \n",
       "66262  @user Europe is racist, but in a different way...   \n",
       "66263  A labour of love, inspired by Middle-earth.\\n\\...   \n",
       "66264  @user I must have missed the great flight of t...   \n",
       "66265  #Wuerzburg today an African immigrant or refug...   \n",
       "\n",
       "                                              scores  Negative   Neutral  \\\n",
       "0       ['0.0679844', '0.88380396', '0.0482117', '']  0.067984  0.883804   \n",
       "1           ['0.3660313', '0.5833739', '0.05059479']  0.366031  0.583374   \n",
       "2          ['0.30297568', '0.4430802', '0.25394407']  0.302976  0.443080   \n",
       "3      ['0.02260253', '0.07604674', '0.9013506', '']  0.022603  0.076047   \n",
       "4          ['0.6982864', '0.27545568', '0.02625783']  0.698286  0.275456   \n",
       "...                                              ...       ...       ...   \n",
       "66261     ['0.03550766', '0.82208896', '0.14240335']  0.035508  0.822089   \n",
       "66262      ['0.8931986', '0.09380516', '0.01299612']  0.893199  0.093805   \n",
       "66263      ['0.15655968', '0.5201203', '0.32332003']  0.156560  0.520120   \n",
       "66264     ['0.91320664', '0.06375945', '0.02303378']  0.913207  0.063759   \n",
       "66265        ['0.92373', '0.06414817', '0.01212183']  0.923730  0.064148   \n",
       "\n",
       "       Positive     CLASS num_CLASS  \n",
       "0      0.048212   NEUTRAL         0  \n",
       "1      0.050595   NEUTRAL         0  \n",
       "2      0.253944   NEUTRAL         0  \n",
       "3      0.901351  POSITIVE         1  \n",
       "4      0.026258  NEGATIVE        -1  \n",
       "...         ...       ...       ...  \n",
       "66261  0.142403   NEUTRAL         0  \n",
       "66262  0.012996  NEGATIVE        -1  \n",
       "66263  0.323320   NEUTRAL         0  \n",
       "66264  0.023034  NEGATIVE        -1  \n",
       "66265  0.012122  NEGATIVE        -1  \n",
       "\n",
       "[38611 rows x 28 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "33387146-7e41-4291-8ce8-95d8dc9f6dff",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9cdcce92-b367-49b7-ba0a-cb3c55d8d86e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>Language</th>\n",
       "      <th>date</th>\n",
       "      <th>inflow</th>\n",
       "      <th>new text</th>\n",
       "      <th>scores</th>\n",
       "      <th>Negative</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>['0.0679844', '0.88380396', '0.0482117', '']</td>\n",
       "      <td>0.067984</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>['0.3660313', '0.5833739', '0.05059479']</td>\n",
       "      <td>0.366031</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>['0.30297568', '0.4430802', '0.25394407']</td>\n",
       "      <td>0.302976</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>['0.02260253', '0.07604674', '0.9013506', '']</td>\n",
       "      <td>0.022603</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>German</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>['0.6982864', '0.27545568', '0.02625783']</td>\n",
       "      <td>0.698286</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>38606</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-28 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>For day 1 of week 2, @user discusses the Socia...</td>\n",
       "      <td>['0.03550766', '0.82208896', '0.14240335']</td>\n",
       "      <td>0.035508</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38607</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-27 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>@user Europe is racist, but in a different way...</td>\n",
       "      <td>['0.8931986', '0.09380516', '0.01299612']</td>\n",
       "      <td>0.893199</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38608</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-27 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>['0.15655968', '0.5201203', '0.32332003']</td>\n",
       "      <td>0.156560</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38609</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-26 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>@user I must have missed the great flight of t...</td>\n",
       "      <td>['0.91320664', '0.06375945', '0.02303378']</td>\n",
       "      <td>0.913207</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38610</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>English</td>\n",
       "      <td>2021-06-25 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>['0.92373', '0.06414817', '0.01212183']</td>\n",
       "      <td>0.923730</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "38606  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "38607  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "38609  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "38610  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "38606  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "38607  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "38608  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "38609  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "38610  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "38606                           0.0                         1.0   \n",
       "38607                           0.0                         1.0   \n",
       "38608                           0.0                         1.0   \n",
       "38609                           0.0                         0.0   \n",
       "38610                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "38606                        2.0                         0.0   \n",
       "38607                        6.0                         0.0   \n",
       "38608                       17.0                         0.0   \n",
       "38609                        1.0                         0.0   \n",
       "38610                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ... Language  \\\n",
       "0                                  0.0  ...   German   \n",
       "1                                  0.0  ...   German   \n",
       "2                                  0.0  ...   German   \n",
       "3                                  0.0  ...   German   \n",
       "4                                  0.0  ...   German   \n",
       "...                                ...  ...      ...   \n",
       "38606                              0.0  ...  English   \n",
       "38607                              0.0  ...  English   \n",
       "38608                              0.0  ...  English   \n",
       "38609                              0.0  ...  English   \n",
       "38610                              0.0  ...  English   \n",
       "\n",
       "                            date      inflow  \\\n",
       "0      2016-04-20 00:00:00+00:00     Syrians   \n",
       "1      2016-04-20 00:00:00+00:00     Syrians   \n",
       "2      2016-04-20 00:00:00+00:00     Syrians   \n",
       "3      2016-04-20 00:00:00+00:00     Syrians   \n",
       "4      2016-04-20 00:00:00+00:00     Syrians   \n",
       "...                          ...         ...   \n",
       "38606  2021-06-28 00:00:00+00:00  Ukrainians   \n",
       "38607  2021-06-27 00:00:00+00:00  Ukrainians   \n",
       "38608  2021-06-27 00:00:00+00:00  Ukrainians   \n",
       "38609  2021-06-26 00:00:00+00:00  Ukrainians   \n",
       "38610  2021-06-25 00:00:00+00:00  Ukrainians   \n",
       "\n",
       "                                                new text  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...   \n",
       "...                                                  ...   \n",
       "38606  For day 1 of week 2, @user discusses the Socia...   \n",
       "38607  @user Europe is racist, but in a different way...   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...   \n",
       "38609  @user I must have missed the great flight of t...   \n",
       "38610  #Wuerzburg today an African immigrant or refug...   \n",
       "\n",
       "                                              scores  Negative   Neutral  \\\n",
       "0       ['0.0679844', '0.88380396', '0.0482117', '']  0.067984  0.883804   \n",
       "1           ['0.3660313', '0.5833739', '0.05059479']  0.366031  0.583374   \n",
       "2          ['0.30297568', '0.4430802', '0.25394407']  0.302976  0.443080   \n",
       "3      ['0.02260253', '0.07604674', '0.9013506', '']  0.022603  0.076047   \n",
       "4          ['0.6982864', '0.27545568', '0.02625783']  0.698286  0.275456   \n",
       "...                                              ...       ...       ...   \n",
       "38606     ['0.03550766', '0.82208896', '0.14240335']  0.035508  0.822089   \n",
       "38607      ['0.8931986', '0.09380516', '0.01299612']  0.893199  0.093805   \n",
       "38608      ['0.15655968', '0.5201203', '0.32332003']  0.156560  0.520120   \n",
       "38609     ['0.91320664', '0.06375945', '0.02303378']  0.913207  0.063759   \n",
       "38610        ['0.92373', '0.06414817', '0.01212183']  0.923730  0.064148   \n",
       "\n",
       "       Positive     CLASS num_CLASS  \n",
       "0      0.048212   NEUTRAL         0  \n",
       "1      0.050595   NEUTRAL         0  \n",
       "2      0.253944   NEUTRAL         0  \n",
       "3      0.901351  POSITIVE         1  \n",
       "4      0.026258  NEGATIVE        -1  \n",
       "...         ...       ...       ...  \n",
       "38606  0.142403   NEUTRAL         0  \n",
       "38607  0.012996  NEGATIVE        -1  \n",
       "38608  0.323320   NEUTRAL         0  \n",
       "38609  0.023034  NEGATIVE        -1  \n",
       "38610  0.012122  NEGATIVE        -1  \n",
       "\n",
       "[38611 rows x 28 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "ea24fc10-a70b-476d-b378-9766940a8b9a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 38611 entries, 0 to 38610\n",
      "Data columns (total 28 columns):\n",
      " #   Column                           Non-Null Count  Dtype  \n",
      "---  ------                           --------------  -----  \n",
      " 0   text                             38611 non-null  object \n",
      " 1   author_id                        38611 non-null  float64\n",
      " 2   created_at                       38611 non-null  object \n",
      " 3   lang                             38611 non-null  object \n",
      " 4   geo.place_id                     38225 non-null  object \n",
      " 5   public_metrics.retweet_count     38611 non-null  float64\n",
      " 6   public_metrics.reply_count       38611 non-null  float64\n",
      " 7   public_metrics.like_count        38611 non-null  float64\n",
      " 8   public_metrics.quote_count       38611 non-null  float64\n",
      " 9   public_metrics.impression_count  38611 non-null  float64\n",
      " 10  geo.coordinates.coordinates      4347 non-null   object \n",
      " 11  in_reply_to_user_id              12325 non-null  float64\n",
      " 12  entities.hashtags                15559 non-null  object \n",
      " 13  week                             38611 non-null  float64\n",
      " 14  month                            38611 non-null  float64\n",
      " 15  year                             38611 non-null  float64\n",
      " 16  year-week                        38611 non-null  object \n",
      " 17  year-month                       38611 non-null  object \n",
      " 18  Language                         38611 non-null  object \n",
      " 19  date                             38611 non-null  object \n",
      " 20  inflow                           38611 non-null  object \n",
      " 21  new text                         38611 non-null  object \n",
      " 22  scores                           38611 non-null  object \n",
      " 23  Negative                         38611 non-null  float64\n",
      " 24  Neutral                          38611 non-null  float64\n",
      " 25  Positive                         38611 non-null  float64\n",
      " 26  CLASS                            38611 non-null  object \n",
      " 27  num_CLASS                        38611 non-null  int64  \n",
      "dtypes: float64(13), int64(1), object(14)\n",
      "memory usage: 8.2+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0e83b294-399d-40dd-869c-4abd19611c03",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NEGATIVE    17702\n",
       "NEUTRAL     17084\n",
       "POSITIVE     3825\n",
       "Name: CLASS, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the unique value counts of CLASS (absolute frequency)\n",
    "\n",
    "df['CLASS'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "46155f16-9cc0-4b99-a62f-4189cb2e737d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NEGATIVE    0.458470\n",
       "NEUTRAL     0.442465\n",
       "POSITIVE    0.099065\n",
       "Name: CLASS, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the relative frequencies of CLASS\n",
    "\n",
    "df['CLASS'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e2037cf7-71e2-4b95-ab11-d691c1e9908e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: ylabel='CLASS'>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb8AAAGFCAYAAABt8v25AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAABCi0lEQVR4nO3dd3hUZcIF8HOnZCaZmfQeEgKkIAnSEUSQ3lVQFBGUasH+WXddXVZxUWyr4urqSrEDAhZYFFSkSJEaQJASWkISSnqbPvf7A43GBEmZzDsz9/yeZx7JvTczZwLmzC3veyVZlmUQEREpiEp0ACIiIk9j+RERkeKw/IiISHFYfkREpDgsPyIiUhyWHxERKQ7Lj4iIFIflR0REisPyIyIixWH5ERGR4rD8iIhIcVh+RESkOCw/IiJSHJYfEREpDsuPiIgUh+VHRESKw/IjIiLFYfkREZHisPyIiEhxWH5ERKQ4LD8iIlIclh8RESkOy4+IiBSH5UdERIrD8iMiIsVh+RERkeKw/IiISHFYfkR+YMqUKZAkCc8//3yt5Z9//jkkSQIArF+/HpIk1fs4c+ZMzfeUl5fjqaeeQkZGBgIDAxEREYEePXrghRdeQElJSZ3X/vjjj6FWq3HXXXfVLOvfv/9FX0uSJCQnJ9ds9+CDDwIAOnbsiBkzZtT7/j755BNotVqcPXu2we+D6M+w/Ij8hF6vx9y5c+stqN87fPgwCgoKaj2io6MBAMXFxejVqxcWLlyIRx55BD/++CM2b96MWbNmISsrCx9//HGd51uwYAEee+wxLF68GNXV1QCAFStW1Dz39u3bAQDffvttzbIdO3bUeZ7p06dj6dKlNc/xx9cYPXo0YmJiGvQ+iC5FIzoAEbnH4MGDkZ2djeeeew4vvPDCRbeLjo5GaGhoveueeOIJ5OTk4PDhw0hISKhZ3r59e4wePRqyLNfa/uTJk9iyZQuWL1+O77//HsuWLcNtt92G8PDwmm0sFgsAICIiArGxsRfNdeutt+Lxxx/Hp59+ismTJ9csz8nJwbp16/DFF180+H0QXQr3/Ij8hFqtxpw5czBv3jycPn260d/vcrmwZMkSTJo0qVbx/d6vh1B/tWDBAowaNQohISGYNGkS5s+f36TswIVyvO6667Bw4cJayxcuXIiYmBiMGDGiyc9N9EcsPyI/MnbsWHTu3BmzZs266DatWrWC0WiseaSnpwMAzp8/j9LS0pqvf9WtW7eabSdMmFCz3OVyYdGiRZg0aRIA4Oabb8bWrVuRnZ3d5PzTpk3Dxo0bcfz4cQCALMtYtGgRpkyZArVa3aD3QdQQPOxJ5Gfmzp2LgQMH4uGHH653/aZNm2AymWq+1mhq/xr4497dZ599BpvNhscffxxms7lm+dq1a1FVVVWzRxYZGYmhQ4diwYIFmDNnTpOyDx06FK1atcLChQsxe/ZsrFu3DidPnsTUqVMb/T6I/gz/tRD5mX79+mHYsGF44oknMGXKlDrr27RpU++5sqioKISGhuLQoUO1liclJQEATCYTSktLa5YvWLAAxcXFCAoKqlnmcrmwZ88ezJ49u86eWkOoVCpMmTIFixYtwtNPP42FCxeiX79+SE1NbfD7IGoIHvYk8kPPP/88Vq5ciS1btjT4e1QqFW666SZ8+OGHyMvL+9Nti4qK8MUXX2Dx4sXIysqq9aisrMRXX33V5OxTp07F6dOnsWLFCqxYsQLTp09v8nMRXQz3/Ij8UMeOHTFx4kTMmzevzrpz587VXIH5q4iICGi1WsyZMwfr16/HFVdcgWeeeQbdu3eHwWDAvn37sHXrVmRmZgIAPvjgA0RERODGG2+ESlX7M/To0aMxf/58jB49uknZ27Rpg4EDB+KOO+6AVqvFuHHj6t3uz94H0aWw/Ij81OzZs7F06dI6y+u7MGTr1q3o1asXIiIisH37dsydOxcvvvgiTpw4AZVKhdTUVIwfP75mQPqCBQswduzYOsUHADfccAPGjx+Ps2fP1hqX1xjTp0/Hd999hzvuuKPWYdWGvg+iS5HkPw7cISIi8nM850dERIrD8iMiIsVh+RERkeKw/IiISHFYfkREpDgsPyIiUhyWHxERKQ4HuRM1kmy1QK6uvvAwV8NVXfW7r6sgm82/fF0F2fzLcqsVklYLKTAQkk4P6PWQdHpI+l8ev/wZNcsCIen1UIWFQxUSKvotE/kdlh9RPVylJXDm5cKZd/qXR27N13JFuWfD6PVQR8VAFRMLdXQMVNExUMfEQZ2QCHViElShYZ7NQ+QHOMMLKZarqhLO48dqFZvzdC6c+achV1WKjtdgktEEdavEC2XYKhGadqnQZnSEKjxSdDQir8XyI8Vwnj8H+/4s2PfvhX1/FpwnjgEul+hYLUYVGwdtxuXQZnSEJuNyaNqlQFLzYA8RwPIjPyXLMpwnjv1Wdj/tg+tsgehYYukDoW3f4Zcy7Ahth448n0iKxfIjvyA7HbAf2A/7viw4ftoH+4F9kCsrRMfyeurEJGgyLkdA1x4I6N0XKqNRdCQij2D5kc+SHQ7Ydm2HbcN3sG7eALncwxei+BuNBtou3aHrNxC6q67mhTTk11h+5FNkux22nT/CunEdbJs3ev7KS6VQqaHNvBwB/QZC168/1FFNuy8fkbdi+ZHXk222C4W34TvYtmzi4UxPkyRo0jtA168/dP0GQp2QKDoRUbOx/MgryTYbbNu3XtjD27IRclWV6Ej0C3XbFOj6DYB++DVQx8SKjkPUJCw/8irOs2dg/vxTWFZ/wXN43k6lRkCfvggccyMCuvYQnYaoUVh+5BVs+/bAvHwJbD9sAFxO0XGokdTJbRE49kboh4yEFBgoOg7RJbH8SBjZZoN13VqYVyyF4+gh0XHIDSSDEfoRoxE45kaeGySvxvIjj3MVF8L8xQqYV66AXFIsOg61BElCQM/e0I+9CQE9e0OSJNGJiGph+ZHH2A8fhHn5EljXfwvY7aLjkIeoExKhHzMOgaOugxQYJDoOEQCWH3mAbd8eVL37Jhz794qOQgJJYeEw3DoN+muuh6ThHKMkFsuPWozj1AlUvfMGbFs2iY5CXkQVnwDD1DuhGzSMh0NJGJYfuZ2zqBDVC9+G5atVvHKTLkrdLhWG2++B7oorRUchBWL5kdu4qiphXvwBqpd9AlgsouOQj9B26grDHfdC2yFTdBRSEJYfNZvscMDy5XJUfbAAcmmJ6DjkowL69odhxt3QJCWLjkIKwPKjZrF8/w2q5/8Hzrxc0VHIH6jU0A8biaCpd3AybWpRLD9qEvtP+1D573/BceiA6Cjkj/R6GKbeicBxEyCpVKLTkB9i+VGjuKqrUPX2G7CsXAHwnw61ME2HTJgeewqa1m1ERyE/w/KjBrNu24zKfz0P17mzoqOQkmgDEHTbNARNuA2SmuMDyT1YfnRJrrJSVM57Gdbv1oiOQgqmSU2/sBeYkiY6CvkBlh/9KeuGdah4dS6v4iTvoNEgaMJtCLp1OiStVnQa8mEsP6qXq7ISla+/COs3X4mOQlSHOrktTI89Be1lGaKjkI9i+VEdtt07UDH3GZ7bI++mUiNw3M0wTLsTkk4vOg35GJYf1ZBtVlS982+YVyzhlZzkM9SJSQj+x/PQtE0RHYV8CMuPAADO/DyUPfUonMezRUchajx9IEyPPgn9wCGik5CPYPkRbLu2o/yZJyCXl4uOQtQsgTdNhOGOeyGp1aKjkJdj+Slc9dKPUPX2G7z7AvkNbZduCP77HKhCw0RHIS/G8lMo2WZFxUtzeDUn+SVVdAyCn5kLbXoH0VHIS7H8FMh5/izKn3oMjsM/i45C1HICdDA9+Bj0I64RnYS8EMtPYez796Js1uOQS4pFRyHyCP2118N478McFE+1sPwUxLzyM1TOewmw20VHIfIoTUZHBP/jeagjo0RHIS/B8lMA2eFA5byXYPlyhegoRMKowiMQ/M+XoW3P84DE8vN7ropylD/5KOz79oiOQiScFGRA8D9fRkDnrqKjkGAsPz/mKitF2aP3wXH0sOgoRN5Dp0PIMy8goGdv0UlIIN4i2U+5iotQ+n8zWXxEf2S1ouzJR2Dd+L3oJCQQ9/z8kLPwPMoeuhvO3FOioxB5L5Uapsefgn7oSNFJSACWn59xnj2D0ofuhiv/tOgoRN5PkmB88DEEXnuD6CTkYSw/P+LMz7tQfGcLREch8imGu+5H0PhJomOQB7H8/IQj9xTKHr4HrvPnREch8klBt82AYeodomOQh7D8/IDj5PELxVdcJDoKkU8LvPEWGO9+UHQM8gCWn49zHDuK0kfuhVxaIjoKkV/Qjx4L40N/gSRJoqNQC+JQBx9mP3IIpQ/NZPERuZFl1Weoenue6BjUwlh+PspZkIeyvzzIG9AStQDzkg9RvfQj0TGoBbH8fJCrsgJlf32Id2YgakFV/3kdFt7v0m+x/HyM7HCg/O+Pw3nqhOgoRP5NllHxwmzYtm8VnYRaAMvPx1S8PAf2PTtFxyBSBocDZbP+Avuhg6KTkJux/HxI1QfzYf16legYRMpiMaPsiYfgPMPJI/wJy89HWL5bg+qF74iOQaRIckkxyp54CK6qStFRyE1Yfj7Avn8vKubOBjgkk0gY54ljKH/mb5CdTtFRyA1Yfl7OmZeLsicfAew20VGIFM++fSsq33hFdAxyA5afF3OVl/0ylq9MdBRyk9dO5SN6/Q48eTSn1vIjVWbcuv8o2m3ajTabdmHEroM4bbHWrH8qOwdpP+xGl6178dnZ2tPYfXGuGJP2H/FIfgIsn38K84olomNQM2lEB6D6yXY7yp96FM7TuaKjkJvsKa/EBwXn0cEQWGv5CbMF1+z5GbfEReGx5HiYNGocrbZAp7rw2XRNYSlWnC3C0svTcdxswQOHT+Dq8BCEazUoszsw58RpLO+ULuItKVblm69C0z4D2g6ZoqNQE3HPz0tV/fffsO/LEh2D3KTS4cTMn4/j5bRkhGpqf+Z87ngeBkWEYla7RHQ0GZAcqMeQiFBEBWgBAEeqzegTGozOwQZcHxMBo1qNU2YLAOCZ46cxNT4arfQ6j78nRXM6Uf7sk7wAxoex/LyQbcc2mJd9IjoGudFfjp7CkIhQXB0eUmu5S5bxTXEp2gXqcdPew+iweQ+G7zqI1ed/m681wxiErIoqlNod2FtRBYvLhTaBemwrrcC+iirc3irG02+HALgK8lH5yvOiY1ATsfy8jKusFBVzn+GVnX7ks7NF2F9Zjb+1aVVn3XmbA1VOF+blFGBgeAiWXJ6OkZFhmHogG1tKL8zbOjA8BONiIjB010Hcf+gE5rVviyC1Co8fPYWX0pOxKO8cev+4H6N2/4xDVWZPvz1Fs65bC/PqL0XHoCbgOT8vU/HCbLiKCkXHIDfJs1jxt+wcLO2UDr267mdNGRc+5AyPDMVdibEAgI6mIOwor8R7+edxZWgwAOCxNgl4rE1Czfe9cCIP/cKCoZEkvHIqHxt6ZOKbolLc+/NxfNs9wwPvjH5VOe9laDMvhyYpWXQUagTu+XkR8xfLYNuySXQMcqO9FdUotDswZOcBxK3fgbj1O7ClrAL/zTuLuPU7EKbVQCNJSAuqfRFMapAepy31D285WmXG8nNFeDw5AVtKK9A71ITIAC2ujQ7HvspqVDg4Ds2jLGZUzH4Sso3DkXwJ9/y8hOPkcVS+9ZroGORm/cKCseEPe2IPHD6BlKBA3JcYC51Khc6mIGT/cgHLr46ZLUjUB9R5PlmW8fCRk3i6XSKMGjWcsgz7L4fIHa4L/3XxkLnHObKPoOrteTDe97DoKNRALD8vINtsKH/2KcBqvfTG5FOMGjUuMwbVWhakUiNco6lZfk9iHO44eAy9Q0zoE2rC98VlWFtYis86t6/zfB8UnEekVovhkWEAgJ4hRrx4Mh87yyrxXXEZ0oP0CNHyf2sRzCuWQNu9J3S9+4qOQg3Aw55eoOq/b8B57KjoGCTIqKgwvJjWGm/kFqD/zp/wYcF5LMhMQa9QU63tztnseO1UAeakJtUs6xpsxMzEGEzcfwRfni/Ga+3beDo+/U7F3NlwFp4XHYMaQJJlHiMRybZ9K8r+8iCv7iTyE9ou3RDy0r8hqbhv4c34tyOQq7SEwxqI/Ix9zy6YF38gOgZdAstPoIoX/wlXcdGlNyQin1L1/rtwFuSJjkF/guUniPWHDbBt2Sg6BhG1BKsVlW/8S3QK+hMsPwFkmxWVb70qOgYRtSDblo2wbtssOgZdBMtPgOolH8KVz0MiRP6uct7LHPzupVh+HuY8dxbVH78nOgYReYAr/zSql/DiF2/E8vOwqrdeAyyWS29IRH6h+qNFcJ4pEB2D/oDl50G2rF2wrv9WdAwi8iSrFZX/fkV0CvoDlp+HyE4nKue9LDoGEQlg+2EDbNu3io5Bv8Py8xDLl8vhPJ4tOgYRCVI57yXIdrvoGPQLlp8HuMpKUbXwHdExiEgg5+lcVC/5UHQM+gXLzwOq3n0LckW56BhEJFj1RwvhPHtGdAwCy6/F2Y8cgmX1F6JjEJE3sFhQtYhHgbwBy6+FVb37JuByiY5BRF7C+s1XHPrgBVh+Lch+5BDsO7aJjkFE3sTpRPUn74tOoXgsvxZU/fEi0RGIyAtZvloJZ1Gh6BiKxvJrIY6cU7BtWi86BhF5I7sNZl75KRTLr4WYF7/Pc31EdFHmlZ/BVVYqOoZisfxagPP8WVi++Up0DCLyZhYzzMs+EZ1CsVh+LcC89CPA4RAdg4i8nPmzT+GqrBQdQ5FYfm7mKiuFeRXH9RHRpclVlTB/vlR0DEVi+bmZecVSwGIWHYOIfIR52WLIZv7O8DSWnxvJ5mqYP+OnOCJqOLmsFOaVK0THUByWnxuZV37GOTyJqNHMSz+CbLOJjqEoLD83ke12mD/9WHQMIvJBrqJCWDeuEx1DUVh+bmLduA6uwvOiYxCRj7J8vUp0BEVh+bmJZc3/REcgIh9m37MTznNnRcdQDJafGziLCmHftUN0DCLyZS4XLGv5IdpTWH5uYP32a8DlFB2DiHyc9WuWn6ew/NzAsna16AhE5Aecebmw788SHUMRWH7N5Mg+AufxbNExiMhPWL5aKTqCIrD8mokXuhCRO1k3rINssYiO4fdYfs0gO52wfLdGdAwi8iNydRXH/HkAy68ZbDu2QS4pFh2DiPyM5Wse+mxpLL9msPKefUTUAuxZu+E8UyA6hl9j+TWRq6oS1s0bRMcgIn8ky7yeoIWx/JrIuv47wGoVHYOI/JT1u69FR/BrLL8msn7Lf5hE1HKcuTlwFuSJjuG3WH5N4KqsgH1flugYROTnbDu2iY7gt1h+TWDftZ3TmRFRi2P5tRyWXxPYftwiOgIRKYB9z07IDofoGH6J5dcEth0/io5ARAogV1XBfmCf6Bh+ieXXSI7j2XAVnhMdg4gUws5Dny2C5ddIPORJRJ7E834tQyM6gK/ZFBwO17S7EXPuDKKydkB/Old0JCLyY46jh+EqLYEqNEx0FL/C8msEp9OJE3mn4XA4cCzQCPQegFCDAfGyC9G5JxGxaxs0ZWWiYxKRP5Fl2Hb+CP3g4aKT+BWWXyOcPXcWjj9ceVVaVYVSAAcjYiANG4Nokwmx5irEHDuMkN07oLLbhWQlIv9h27GN5edmLL9GyMv789kWZFnG2fJynAWApBRo27ZHbFAQYsuKEXVgH0w//+SRnETkX+w7f4Qsy5AkSXQUv8Hya4S8/MZNNWR3OJBbXo5cSQNkdkVQ9ysRr9Ug5mwBIvfshD6f5wuJ6NJcxUVwHjsKTUqa6Ch+g+XXQE6nE2fOnGnWc1RbLMi2ANlBJqDPAIQZjYhzOhF9+iQidmyDpoLnC4mofvb9WSw/N2L5NdC58+fqnO9rrpLKSpTgwvlC1cixiDIaEWeuQtTRwwjdsx0qzuxARL9wHDsqOoJfYfk1UFFRUYs+v8vl+u18YXIKtKmXIS4wELGlxYg8sBemQwda9PWJyLux/NyL5ddALV1+f2S325FjtyNHpQE6doOh51WIU6sRczYfkVk7oc8/7dE8RCSW48QxyE4nJLVadBS/wPJroKJiz5bfH1WZzcgGkG0IBvoM/OV8oQMxuScRvmMrNJUVQvMRUQuzWuHMy4UmKVl0Er/A8msgT+/5XUrN+cLIWKhG3YBooxGx1VWIOnoIoVk7eL6QyA85jh1l+bkJy68BKqsqYbVaRce4KJfLhTPl5TgDAG1SEZDWAbGBesSWFCPqwF4YDx8UHZGI3MB57CgwYIjoGH6B5dcA3rbXdym2X88XqrXA5d1huKIv4tWqC+MLd2+H7ky+6IhE1AS86MV9WH4N4Gvl90dVZjOOAjhqCAb6Dka40Yg4px3Rp04ifNc2ni8k8hEsP/dh+TWA6Itd3K24shLFAA5Ex9WcL4yrrkTU0Z8RsnsXVC6eLyTyRq7z5+AqL4MqOER0FJ/H8muA4qJi0RFaTO3zhekISMtEXKAeMSVFiNqfBePRQ6IjEtHvOI5nI6BzN9ExfB7L7xJcLheKS/y3/P7IZrfjlN2OU+oAoHNPGHpfjXi1GjFn8hC5+0fozjZvijciah5H9lGWnxuw/C6hrKwMTqdTdAxhas4XGkOAfkMvnC902BGTcwJhO7dBU1UpOiKRojh53s8t3FJ+DocDFosFRqPRHU/nVfztfF9z/Xa+MB6q0eMQYzQitroS0YcPIjhrN88XErUwR+4p0RH8gqoxG69evRoffPBBrWX//Oc/YTQaERoaiqFDh6KkpMStAUUrLy8XHcFruVwuFJSXY4/DhTXt2uOLmydj2/R7cHLszahKTRcdj8gvuUr4gdwdGlV+L730Uq0y2LJlC/7+97/jqaeewtKlS5Gbm4vZs2e7PaRI1dXVoiP4DJvdjlPlFfhRE4BVna/A/ybfhd1TZyJ/+LWwRseKjkfkF1zFyrkGoSVJsizLDd04Ojoaa9asQZcuXQAADz30EA4ePIivv/4awIU9wwceeABHj/rPMelvvv0Gh48cFh3DL0T8Or7w5DGE7foRmqoq0ZGIfFLk6g2QAgNFx/BpjTrnV1FRgYiIiJqvf/jhB4wbN67m64yMDOTn+9fsIdVm7vm5S1FlJYoAIKYVVKOTEGM0Iq6qAlGHDyJ47x6eLyRqIFdJMdSBCaJj+LRGlV98fDx+/vlnJCUlobKyEnv37sW//vWvmvVFRUUICgpye0iRzNVm0RH80q/nCwsAIOUy6Dp0Qpxeh5iiQkT9tAeG7COiIxJ5LVdxEdTxLL/maFT5jRs3Dg8++CCeeOIJrF69GrGxsejVq1fN+p07dyI93b8udOCen2dYbTactNlwUqsDuvSCqc9AxKkkxOSfRsTu7dCdPys6IpHX4EUvzdeo8ps1axby8/Nx//33IzY2Fh9++CHUv7ux4ieffIJrrrnG7SFFkWUZFotFdAxFqqiuRgWAI8FhQP9hiDCZEGe3XjhfuPNHaPihhBTMpaCJN1pKoy54URqLxYJ3F7wrOgb9gVqtRozBgNjKCkQfOYDgvbshuVyiYxF5TNCU22GYfLvoGD7NLYPcN2zYgKqqKvTu3RthYWHueEqvwGEO3snpdCK/vBz5AJDSAboOnRGn0yG2+Dwi9+2B4bj/XG1MVB8XJ99otkaV34svvojKyko8/fTTAC4cFhwxYgTWrl0L4MJQiO+++w4ZGRnuTyqA2cyLXXzBb+cL9UC33jD1HYQ4SUJMQS4id21HQOE50RGJ3IqHPZuvUYPcP/nkE3To0KHm62XLlmHjxo3YtGkTCgsL0b1795pi9AcsP99UUV2NI1VV2BQcjs8GDMd30+7BwUkzUHjVADj1HBtFvo8D3ZuvUXt+J06cwOWXX17z9erVq3HDDTegT58+AIAnn3wSN954o3sTCsQrPf1DYUUFCgEgLhHqMcm/nC8sR/ShAwjev4fnC8nncM+v+RpVfna7HTqdrubrrVu34oEHHqj5Oj4+HoWFhe5LJxj3/PxPrfOFaRnQd+yKuIAAxBSdR9S+XQg6cUx0RKJLksv8aw5lERpVfikpKdi4cSPatm2LnJwcHDlyBFdffXXN+tOnT9eaAcbXKflWRkphsVpxwmrFiQA90L0PTP2GIF4CYvJzEbFrGwKKeGEBeR/ZZhcdwec1qvxmzpyJe++9F5s2bcK2bdvQu3fvWucA161bVzPvJ5EvqqiuxmEAh0MiIA0afWE+UpsV0SeOImzXdqgtPBpAXoAfzJutUeV35513QqPRYNWqVejXrx9mzZpVa31+fj6mTp3q1oBEosiy/Nv5wvjWUCe2RazBgNiKMkQfOgDT/j2QOEyWRHCx/JrL7YPcz58/j6ioKHc+pTBbtm7B7j27RccgL6XX6RAXEIDYwnOI3LsTQadOiI5EChL57RZIarcM1VYkt/zkZFnGV199hfnz52PVqlWwWq3ueFoir1ZzvlAXCPTsi+D+wxAvAdF5uYjYzfOF1MKcToDl12TN+skdP34cCxYswHvvvYfKykqMGjUKixcvdlc2Ip9SXl2NcgCHQi+cL4w0GhFnsyDqeDbCdv0ItZXzxJIb8bxfszS6/CwWC5YtW4Z3330X27Ztw5AhQ1BQUICsrCxkZma2REYinyPLMs5XVOA8ACS0hiapLTQxVnyrXguLzCMj1Hwfq2X41w3kPKtR5Xf33Xdj8eLFSE9Px6RJk7B8+XJERERAq9VCpWrUZDFEiuJwOuHI16BX2BB8E7waRyt5v0JqHom/c5ulUeX3zjvv4PHHH8df/vIXmEymlspE5Lc0JVoML7sWaW2PYHXJSsjg1aLUNGpJfemN6KIa9dHh/fffx/bt2xEXF4fx48dj1apVcDgcLZWNyC9JLgntstMxQzcTEbpI0XHIR7H8mqdR5XfLLbfgm2++wU8//YT27dvjnnvuQVxcHFwuFw4ePNhSGYn8UmC+ARNKJqNnaC/RUcgHqVUsv+Zo1jg/WZaxZs0aLFiwAF9++SUiIyNx/fXX4/XXX3dnRmE4zo885VybAnxW+SmsLl4MQ5emltT4+abjomP4tGadMZUkCcOHD8fSpUuRn5+PRx55BBs3bnRXNuEkSRIdgRQi+kQcpjtnoq2hnego5AOCNAbREXye20ZIhoeHY/LkyQgPD3fXUwr3+ztYELU0TbEWo8rG4mibn/F1yf9ExyEvFqYLEx3B57n1WtmcnBy/mtszKIijaMizJKeEtOwOmB54F8IC+AuO6hcW4D87GaJwoMifCAzkXb9JDMNpEyaWTUW3kB6io5AX4p5f87H8/kRQIPf8SBxVtRpXHr8aN4XcAq0qQHQc8iJhOu75NRfL709wz49EkyAh9ngCZrhmonVQsug45CW459d8jbrg5VJDGPLy8poVxtuw/MhbaIsCcG3ZOBxK/gnflHwtOg4Jxj2/5mtU+f3rX/+65DZJSUlNDuNt1Go1dDodb9FEXkFySLgsuyNaJSZhhX0pyuyloiORILwYqvkaVX4nTijvZp2BgYEsP/IqptwQ3GqYhh9i1iOrjJMwKBH3/JqvUef81q1bhw4dOqC8vLzOurKyMmRkZGDTpk1uC+cNeNELeSNVlRr9TgzCuJDx0Ei8oanSsPyar1Hl9+qrr+L2229HcHBwnXUhISG488478corr7gtnDfgeT/yWjIQfzwJM6S70SooUXQa8iBe8NJ8jSq/vXv3Yvjw4RddP3ToUOzatavZobwJB7qTtws4r8PYM+MxMGyw6CjkIVH6KNERfF6jyu/s2bPQarUXXa/RaHD+/Plmh/Im3PMjXyDZJWRmd8EUw+0waXivTX8WrotAcECI6Bg+r1Hll5CQgP379190/b59+xAXF9fsUN7EaDCKjkDUYME5oZhcfTs6Bl8uOgq1kLbBnPzcHRpVfiNHjsTf//53WCyWOuvMZjNmzZqF0aNHuy2cNwgL57F18i2qCjX6nxyKsaE38oanfqitieXnDo26n9/Zs2fRtWtXqNVq3HvvvUhPT4ckSfj555/x73//G06nE7t370ZMTExLZvYom82Gd959R3QMoiaxRpuxMuAz5Jv9awIKJftL5ycxLf120TF8XqOukY6JicGWLVswc+ZM/PWvf8WvvSlJEoYNG4Y333zTr4oPAAICAmAymVBRUSE6ClGj6c4F4oaAm7EvaTc2lHwvOs5FmX+2oGRlOSwn7HCWOBH3cCSMPX672EyWZRQvK0PZuiq4Kl3QpwQgaloYdIm/zXl6/v0SlG+ogkovIXJiKExX/nbPu4qtVajYVIX4x6I9+r5aAvf83KPRA4Rat26N1atXo6SkBNnZ2ZBlGampqQgL89/DgxERESw/8lmSTYVO2d3RunUbLLcsQZWjSnSkOlwWGQGtAxDc34iCVwrrrC/5sgKlqysQMzMC2jgNileUI2/OeSS/EgdVoAqVu6pRsbkKCU9EwX7GgbNvFSOoox5qkxrOKheKlpQh4UnfLz4AaBecIjqCX2jyxNZhYWHo0aMHevbs6dfFBwAR4RGiIxA1W+ipCEwx34HLTBmio9Rh6BKIyPGhMPasO7RIlmWUflWOsDEhMPYMgi4xADF3R0C2ulCx+UKR2/IcCOygh76dDqY+BqiCJNjPOQAAhR+VIGSIEdpI358MQK/WI8HQSnQMv8C7OjRARATLj/yDulyDITkjcV3o9VBJvvG/v+OcE85SF4Iu19csU2klBF6mh/mIDQCga62F9bgNzkoXLMdtkG0ytDFamA9ZYD1pR+gI/xj+0drYxmf+3ryd738U8gDu+ZFfcQGtj7XDjNiZ+EK9HGctZ0Qn+lOOUicAQBNS+5e+OkQFR+GFdYZOgbBcZUXu385ACpAQMzMCKr2Ec/NLEDMzAmVrK1G6pgJqkwrRt4fXOlfoSzjMwX1Yfg0QGhoKlUoFl8slOgqR2+jPBOEm3UTsTtyOzSU+MCevJNWz7Lc/RtwYiogbQ2u+Lvq0FEGZekhqoPizMiS9GIeq3WacfbMISc/55nhklp/7cP+5AdRqNUJDQ0XHIHI7yapCt+xemGSagkC1d07lpwm9MFbx1z3AXznLXFCH1D+O0ZZnR8XmakSMD4H5oBWBl+mhCVbD1CsI1hN2OKt984MsL3ZxH5ZfA/G8H/mz8JNRmGa7E2mm9qKj1KGJVkMdqkL1/t8m15AdMsw/WxCYVvfwpSzLOPvfYkROCoVKr4LskiE7LwzLkn/tzwaPbvYuncI7i47gN1h+DcTzfuTv1KUaDMsZjWvCxkBCPYcYW5DL4oL1pA3WkxcuYLGfc8B60gZ7oQOSJCF0RDBKPi9D5fZqWHNtOPNmESSdCqY+hjrPVf5dFTTBKhi7X9iTDUzXwfyTBeajVpSuLkdAKy3UBt/71RcTGItEo//cLFw0nvNrIO75kRJILgltslMxI+5ufCYtRaHVMxPVW47ZkDf7XM3XhR+UAgBM/QyIvTsCYdeaINtcOLegGK4qF/QpOiQ8EQVVYO0Sc5Q6Ufx5GRKf+W2yDX2KDqGjg5E/9zzUwSrE3u2b/y93i+whOoJfadT0ZkpmNpsxf+F80TGIPMald2FnwjZsK90sOgoB+HvXZzApdbLoGH7D9/b9BQkMDERkRKToGEQeo7Ko0PPYlbgl+Dbo1fpLfwO1qO5RPUVH8Cssv0Zo1YozK5DyRJ6IwTT7XUgxpoqOoljB2mCkhaSLjuFXWH6NkJiYKDoCkRCaEi1GnL4OI8KuER1FkbpGdufMLm7Gn2YjxMfFQ6Xij4yUSXJKSM1ujxm6mQgP8M2LRnxVtyhe7OJu/E3eCFqtFrGxsaJjEAkVlG/ExNIp6BF6hegoitGdV3q6HcuvkRJb8dAnkWRWodexvrg5ZBICVL45T6av0Kl16BjeSXQMv8PyaySWH9EFEiREH4/DdNdMtDG0FR3Hb10e3hkBan7AcDeWXyNFR0cjIID/EIl+pS0KwOj86zE0bIToKH7pqti+oiP4JZZfI6lUKiQkJIiOQeRVJIeE9tmZmBZ4J0IDQkXH8SuD4oeKjuCXWH5NwEOfRPUzng7GpPJp6BrSTXQUv5BoSEJaKMf3tQSWXxOw/IguTlWlRp/jA3BTyARoJa3oOD5tUMIQ0RH8FsuvCcLCwhAcHCw6BpHXkiAh9ngrTMdMJAW1Fh3HZ7H8Wg7Lr4nSUtNERyDyegGFOowpuBGDw4aJjuJzQgPC0D2y6fN5TpkyBZIkQZIkaLVatG3bFo888giqqqpqtnnvvffQs2dPGAwGmEwm9OvXD6tWrarzXG+//TY6deoEg8GA0NBQdOnSBXPnzq1Z/49//AOdO3cGACQnJ9e8bn2P/v3712z36quvwmazITIyEs8++2y97+O5555DZGQkbDYbFi1aVO9z6vWNn3uW5ddE6Wk8Dk/UIA4JHbIvx9Sg2xGsDRGdxmcMaTUMalX9d6pvqOHDh6OgoADHjx/Hs88+izfffBOPPPIIAOCRRx7BnXfeiZtuugl79+7F9u3b0bdvX1x33XV44403ap5j/vz5eOihh3D//fdj79692Lx5Mx577DFUVlbW+5o7duxAQUEBCgoKsHz5cgDA4cOHa5atWLGi1vYBAQGYNGkSFi1ahPpuMrRw4ULceuutNVfZBwcH1zzXr49Tp041+mfD+/k1UVhYGKKjo3Hu3LlLb0xEMOWG4jbjdGyMWod95Vmi43i9kYmjm/0cOp2uZlaqW265Bd9//z0+//xzTJ48GS+//DJef/113HfffTXb//Of/4TFYsFDDz2E6667DomJiVi5ciVuuukmTJ8+vWa7jIyMi75mVFRUzZ/Dw8MBXBgiFhoaetHvmT59Ol577TVs3LgRV199dc3yTZs24ejRo7VeW5Ikt8y0xT2/ZuDeH1HjqCrV6H9yCMaFjodG4mfviwnThaNX9JVuf97AwEDY7XZ88sknMBqNuPPOO+ts8/DDD8Nut9fstcXGxmLbtm1N2rtqqI4dO6JHjx5YuHBhreULFixAz549kZmZ6fbXZPk1Q2pKKie6JmosGYg/loQZqplICORtwuozJKH5hzz/aPv27fj4448xaNAgHDlyBO3atat3wo74+HiEhITgyJEjAIBZs2YhNDQUycnJSE9Px5QpU7B06VK4XC635ps2bRqWLVtWczi1srISn376aa29PgAoKyuD0Wis9Rg6tPFjIfmbuxmCgoI47IGoiQLO6XH92fEYEDZIdBSvM8INhzwBYNWqVTAajdDr9ejduzf69euHefPmXfL7ZFmGJEkAgLi4OGzduhX79+/H/fffD7vdjsmTJ2P48OFuLcAJEybA5XJhyZIlAIAlS5ZAlmXcfPPNtbYzmUzIysqq9fjjHmND8LhDM6Wnp+NUTssdDiDyZ5JdhY7ZXZGU1AbLrYtR6aj/IgolSQhqhd4x7jnkOWDAALz11lvQarWIj4+HVnth3GVaWhp++OEH2Gy2Ont/+fn5KC8vR2pq7ZsXZ2ZmIjMzE/fccw9++OEH9O3bFxs2bMCAAQPckjUkJATjxo3DwoULMX36dCxcuBDjxo2rM6xMpVIhJSWl2a/HPb9matumbc0/KCJqmpCcMEw234GM4I6iowh3U7sJbrtxrcFgQEpKClq3bl3r99TNN9+MyspKvP3223W+56WXXoJWq8UNN9xw0eft0KEDANQaNuEO06dPx+bNm7Fq1Sps3ry5ziFPd+KeXzNpNBq0a9cOhw4dEh2FyKepy9UYWDEMaW3T8WXZZ3DKTtGRPE6r0uLGtuNb/HV69+6NBx54AI8++ihsNhvGjBkDu92ODz/8EK+99hpeffVVJCZeOKUzc+ZMxMfHY+DAgWjVqhUKCgrw7LPPIioqCr1793ZrrquvvhopKSm47bbbkJKSgn79+tXZRpZlnDlzps7y6OjoRl2DwT0/N+BVn0TuIckSEo+1wQz1TMQFxouO43FDEoYjUh916Q3d4NVXX8Wbb76JxYsXo2PHjujWrRs2bNiAzz//vNbwh8GDB2Pbtm248cYbkZaWhhtuuAF6vR7fffcdIiIi3J5r2rRpKCkpwbRp0+pdX15ejri4uDqPxg47k+T6RhVSo8iyjEXvL3L7IQAiJZMDXNibtAsbS9aLjuIxHw5Ygp7RvUTHUATu+bmBJElon95edAwivyLZVOic3QO3mqYiSGMQHafFpQSnsvg8iOXnJh0zO3LMH1ELCDsZiamWO3CZqYPoKC3q5nYTRUdQFP62dhOj0YiUds2//JaI6lKXaTA4ZySuDR3rtishvUmQJghjky9+dSW5n//9KxLo11nNicj9JJeE5GMpmKGZiWh9jOg4bjUq6VqYAnibNE9i+blRdFQ04uOVd4UakSfpzwThpsJJuDL0KtFR3OaWdpNER1Aclp+bdenURXQEIr+nsqrQ/VhvTDRNRqA6UHScZukc0QUZ4Rzc72ksPzdLTk5GWGiY6BhEihBxMhrTbHcizei7Y23v7nC/6AiKxPJzM0mS0LVrV9ExiBRDXarFsNxrMCrsWkiQRMdplE7hXdA/fqDoGIrE8msB6WnpMJlMomMQKYbkktAuOx3TdXchUhcpOk6D3Zf5oOgIisXyawEqlQpdu3Dvj8jTgvKNmFAyGVeEunfOyZbQJaIr+sX1Fx1DsVh+LaTDZR1gMPj/rBRE3kYyq3DFsaswIfhW6FQ60XEu6v7Mh0RHUDSWXwtRq9Xo3Kmz6BhEihV1IhbTnHehrcH7Jp/oGtkdfWL7io6haCy/FpSZkcm9PyKBtMUBGJU3BsPDRomOUsv9mf8nOoLisfxakFarRe9e3n/ugcifSU4JadkdMF1/F8ICwkXHQfeonrgyxn8G6Psqll8LS09LR0yMf03FROSLDHkmTCybiu4hPYXmuD+De33egOXXwiRJQt8+PLZP5A1U1Sr0Pt4PN4XcAq0qwOOv3zOqF3rFXOnx16W6WH4eEBsby7u9E3kJCRJijydghmsmkg1tPPa6akmNJ7o85bHXoz/H8vOQ3r16Q6vRio5BRL/QFgXgmvwbMCRsuEdeb0K7SegQlumR16JLk2RZlkWHUIodO3fgx+0/io5BRH9Q2aocK5xLUGorbZHnj9RHYc2IdbxtkRfhnp8HdencBSYjpz1rqrVr1uK+e+7D8mXL612/+OPFuO+e+/D9uu9rLV+xfAUef/Rx/P3Jv2PXzl211u3etRtvv/V2i2Um32A8HYxJ5dPQOaRlZmZ6rNMTLD4vw/LzII1Ggyt782R3U5w6dQqbN29GfEL990vcu3cvTp48iZCQkFrL9+/fj507duKee+/BtWOuxUcffoSqyioAQHV1NVatXIUbx9/Y4vnJ+6mq1Oh3YhDGhdwMreS+UxQ9oq7AmOTr3fZ85B4sPw9LTU1FXFyc6Bg+xWqx4r1F72HCLRMQFBRUZ31paSmWLV2GyVMmQ61W11p39sxZpKalIql1Erp37w69Xo/CokIAwBeff4Gr+l2F8HDxY7/IS8hA/PFEzMBMJAYlNfvpNJIGs7rOdkMwcjeWnwB9r+oLSfKtW6+ItHTpUmRkZKB9+/Z11rlcLrz/3vsYNHgQ4uLrfqhISEhAzqkcVFdXIycnB3a7HVFRUTiWfQy5ubno37+/B94B+RptoQ5jztyEQWFDmvU8t6VNRVoor/T2RhrRAZQoOioaGR0y8NOBn0RH8Xq7du5Cbm4uHn3s0XrXf/vNt1Cr1Li6/9X1rr+sw2Xo0bMHXpz7IrQBWky6dRICAgKwZMkSTLp1EjZt3ISNGzbCYDRgwoQJ9RYoKZNkl5CR3RmJSclYbluCCnt5o74/JjAW93FAu9finp8gfa7sg5DgkEtvqGAlJSVYvmw5bpt8G7TauudgcnJysP779Zh066Q/3ZMeOWokZj09C0/87Ql06twJa9esRXp6OtRqNdZ8vQYPPvQgrrzySnzw/gct+XbIRwXnhGJy5Qx0DO7UqO/7a+enYNBybl9vxaEOAhWcKcCKz1aAfwX127t3L959512oVL99RnO5XJAkCZIk4drrrsUXn39Rq/h+XR8WFoanZz9d5znPnDmDd/7zDh7/y+PYunUrjh87jmkzpsFqteKRhx7BCy+9gMDAQI+8P/ItsiQjv20OvihbAYfs+NNtr4rthwVX88OUN+NhT4HiYuPQrWs37Ny1U3QUr5Seno6//u2vtZZ99MFHiImJweChgxESHILLOlxWa/2bb7yJHj17oFfvXnWeT5ZlLP54McZePxY6vQ4u2QWnywkAcDqdNdsQ1UeSJSQca40Z0TOxMmAF8sx59W4XrA3GnB4veDgdNRbLT7CePXoiJycH586fEx3F6+j1esTH1x7aEKALgMFoqFluMNY+rKRWqxEcHFzvZOJbNm+ByWRCx8s7AgDatm2Lr/73FU6cOIGDBw4iNja23qtJiX4v4Jwe1wdMwL6kXdhQ8n2d9f/o9ixig3ju2NvxnJ9gKpUKQwYPgUbDzyEtqby8HGvXrMW4G8fVLEtOTsbAQQPxn7f+gz2792DirRMFJiRfItkkdMrujtuM02HQ/PYBbFTStRjd+jqByaiheM7PS+zbvw8bN20UHYOIGskZ7MD68G9Q5CjCquFrEBzAC9l8AcvPS8iyjJWrViInN0d0FCJqLBUwYEx/ZMRy4mpfwcOeXkKSJAwcOBA6nU50FCJqpO5durP4fAzLz4sYDUYMuHqA6BhE1Ajx8fHo2UPs3eGp8Vh+XiYlJYU3viXyEXq9HkMHD601FpV8A//GvFD/q/sjMiJSdAwiuoTBAwfDaDSKjkFNwPLzQlqtFqNGjkJQIMecEXmrLp27IDk5WXQMaiKWn5cymUwYOWJknVv0EJF4bZLb8N6cPo7l58ViY2MxcMBA0TGI6HeioqIwdMhQ3pbMx7H8vFx6Wjq6de0mOgYRATAajRg9cnS9dxkh38Ly8wG9ruiFtm3aio5BpGharRajR42GwcDbFPkDlp8PkCQJQwYPQWQkrwAlEkGlUmH4sOG8CtuPsPx8hFarxagRo3jXASIB+l3VD62TWouOQW7E8vMhJpMJI4fzClAiT+rcqTMyMzl1mb9h+fkYXgFK5Dlt27ZFnyv7iI5BLYDl54PS09LR/+r+omMQ+bWY6BgMHcwhDf6K5eejMjMy0b9ff9ExiPxSaGgoRo0cxZtM+zH+zfqwzMxMyJCxYeMG0VGI/EZ4eDjGXDuGF5f5OZafj+uY2RGyLPMu8ERuEBERgTHXjkFgYKDoKNTCWH5+4PKOl0OWZWz6YZPoKEQ+KyoqCtddcx30er3oKOQBLD8/0enyToAMbNrMAiRqrJjoGFx7zbXQ6XSio5CHsPz8SKdOnSBDxg+bfxAdhchnxMbG4trR1yIgIEB0FPIglp+f6dypM2RZxuYtm0VHIfJ68XHxGD16NAK0LD6lYfn5oS6du0CWZWzZukV0FCKv1SqhFUaNHMU7NCiUJMuyLDoEtYzDhw9j3fp1cDqdoqMQeZXExESMGsFxfErG8vNzBQUF+N9X/4PFYhEdhcgrpLRLwZDBQzhHrsKx/BSgrKwMq1avQklJiegoREJd0fMK9OjeQ3QM8gIsP4WwWq34es3XyD2dKzoKkcdptVoMGTQEbdvyptB0ActPQVwuFzZs3IADBw+IjkLkMSaTCaNGjuKNaKkWlp8CZe3NwuYtm8G/evJ38fHxGDFsBKcrozpYfgp14uQJrP1mLex2u+goRC0iMyMTfa/qywtbqF4sPwUrLCzEqtWrUFlZKToKkduoVCr0vaovOmZ2FB2FvBjLT+Gqq6vx3fff4dSpU6KjEDWbXq/H8GHD0Sqhlego5OVYfgQA2P/TfmzeshkOh0N0FKImiY6OxrAhwxASEiI6CvkAlh/VKC4uxjfffoPzhedFRyFqMJVKhe7duqN7t+5QqVSi45CPYPlRLU6nE9t+3IasvVm8GpS8XnhYOAYPHozoqGjRUcjHsPyoXqfzTuPbb79FZRUvhiHvI0kSOl3eCb2u6MX5OalJWH50URaLBes3rEf2sWzRUYhqBAcHY9DAQUiITxAdhXwYy48u6dChQ9iwaQPHBJJwHTp0wFV9ruL996jZWH7UIGVlZfh+w/c4ffq06CikQEFBQRg4YCCSWyeLjkJ+guVHjXLs2DH8sOUHVFRUiI5CCpGamop+ffshUM8pysh9WH7UaA6HA3uy9mDX7l0cF0gtJjo6Gn379EVcXJzoKOSHWH7UZBUVFdi8ZTMviCG3MhqM6N2rN9LS0iBJkug45KdYftRseXl52PjDRhQVFYmOQj5Mq9Gia9eu6NK5C4cvUItj+ZFbuFwuHDh4ANt+3Aar1So6DvkQSZLQPr09el3RCwaDQXQcUgiWH7mVxWLBj9t/xE8HfuIMMXRJCfEJuKrPVYiKihIdhRSG5UctoqSkBDt37cSRo0dYglRHSEgI+vTug7Zt24qOQgrF8qMWVVZWhp27duLwkcNwuVyi45BgkZGR6NK5C1JTUjkJNQnF8iOPKC8vx67du/DzoZ9ZggqUlJiELp27IDExUXQUIgAsP/Kwquoq7Nu3Dz8d+IkXxvg5lUqF1JRUdOncBZGRkaLjENXC8iMhbHYbfj74M7L2ZXG2GD8TEBCADpd1QOdOnWE0GkXHIaoXy4+EcrlcyD6WjQMHDiAvP090HGoGg8GATpd3QkaHDOh0OtFxiP4Uy4+8Rnl5OQ4fOYxDhw+hrKxMdBxqAJVKhcRWiUhPT0e7tu2gVqtFRyJqEJYfeaWCggIcOnwI2ceyeW7QC8XGxiItNQ2pKakIDOSE0+R7WH7k1RwOB06cPIFDhw4hJzeHYwYFCgsLQ3pqOtLS0hAcHCw6DlGzsPzIZ1RXV9ccFuU8op5hMBiQlpqGtLQ0REVyFhbyHyw/8kllZWXIyc1Bbm4uTuedhs1mEx3JbxiNRrROao3U1FQkxCfwzgrklzjFQguaMmUKJEnC888/X2v5559/XvMLZf369ZAkqd7HmTNnap5nzJgxdZ4/KysLkiTh5MmTNa/1Z4/fZ5IkCRqNBklJSZg5cyZKSkrqPL/ZbEZYWBjCw8NhNpvrrE9OTsarr77azJ9S04SEhKBjZkeMHDESM6bNwPVjr0eP7j0QExPDX9aNpNVo0bp1a/S9qi8mTpiIKbdNwYD+A9AqoRV/luS3eN+QFqbX6zF37lzceeedCAsLu+h2hw8frnMeJTo6usGv89prr9Uq2bi4OCxcuBDDhw+vs+3w4cOxcOFCOBwOHDx4ENOmTUNpaSk++eSTWtstX74cmZmZkGUZK1aswMSJExucx5NUKhXi4+IRHxePK3peAYvFgtN5p2v2DDmOsDa1Wo3YmFjEx8cjISEBcbFxvEqTFIfl18IGDx6M7OxsPPfcc3jhhRcuul10dDRCQ0Ob/DohISEICQmptSw0NBSxsbF1ttXpdDXLW7VqhfHjx2PRokV1tps/fz4mTZoEWZYxf/58ry2/P9Lr9Uhpl4KUdikAgJLSEuSdzsP5wvMoLCpEcXEx7Ha74JSeo9PpEB0djYT4BMTHxyMmOoZlR4rH8mtharUac+bMwS233IL7778frVq1Eh2pluPHj+Prr7+GVquttfzYsWPYunUrVqxYAVmW8eCDD+L48eM+OQt/WGgYwkJ/2+uWZRll5WUoLCxEUVERCgsLUVhU6PN7iDqdDuFh4QgPr/0wBPEeeUR/xPLzgLFjx6Jz586YNWsW5s+fX+82fyzFhIQEHD58uEXyrFq1CkajEU6nExaLBQDwyiuv1NpmwYIFGDFiRM2h2uHDh2PBggV49tlnWySTJ0mShNCQUISGhNbsHQKA1Wq9UIZFhSgsLER5eTmqqqtQXV3tVWMNdTpdzbnYiLAIhIeHIyw8DEYDpxIjaiiWn4fMnTsXAwcOxMMPP1zv+k2bNsFkMtV8rdG03F/NgAED8NZbb6G6uhrvvvsujhw5gvvuu69mvdPpxHvvvYfXXnutZtmkSZPwf//3f3j66af99pCZTqdDfHw84uPj66xzOp2orq6uefxair//2mw2w+l0QpblWg8AdZb9/iJrnU6HwMBA6PV6BOp/+W9g4IU/B+prf63X19lLJ6LGY/l5SL9+/TBs2DA88cQTmDJlSp31bdq0ueg5v+DgYJw6darO8tLSUgCoc67vUgwGA1JSLuzxvP766xgwYACefvppzJ49GwCwZs0a5OXlYfz48bW+z+l0Yu3atRgxYkSjXs8fqNVqmEymWh9QiMh3caiDBz3//PNYuXIltmzZ0qjva9++PX766aeaQ5S/2rFjB6Kiov70KtKGmDVrFl566SXk5+cDuHChy80334ysrKxaj4kTJ170sC0RkS/hnp8HdezYERMnTsS8efPqrDt37lydcouIiIBWq8XEiRMxe/Zs3HrrrXj88ccRFhaGrVu34rnnnsNf//rXZufq378/MjIyMGfOHMyaNQsrV67El19+iczMzFrbTZ48GaNGjcL58+cRFXVhto+8vDxkZWXV2i4pKQnh4eHNzkVE1FK45+dhs2fPrnd+yvT0dMTFxdV67Nq1C8CFw5qbNm2CLMsYM2YMOnXqhBdeeAGzZ8++6DnExnrooYfw3//+F2+++SYMBgMGDRpUZ5sBAwbAZDLhgw8+qFn20ksvoUuXLrUeX375pVsyERG1FE5vRkREisM9PyIiUhyWHxERKQ7Lj4iIFIflR0REisPyIyIixWH5ERGR4rD8iIhIcVh+RESkOCw/IiJSHJYfEREpDsuPiIgUh+VHRESKw/IjIiLFYfkREZHisPyIiEhxWH5ERKQ4LD8iIlIclh8RESkOy4+IiBSH5UdERIrD8iMiIsVh+RERkeKw/IiISHFYfkREpDgsPyIiUhyWHxERKQ7Lj4iIFIflR0REisPyIyIixWH5ERGR4rD8iIhIcVh+RESkOCw/IiJSHJYfEREpDsuPiIgUh+VHRESKw/IjIiLFYfkREZHisPyIiEhxWH5ERKQ4LD8iIlIclh8RESnO/wMouAdTmbqEmgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# pie chart\n",
    "\n",
    "# first color: greyish hex code\n",
    "# second color : redish hex code\n",
    "# third color: greenish hex code\n",
    "\n",
    "colors = '#EA4035', '#939593', '#25A81E'\n",
    "\n",
    "# adding percentages\n",
    "\n",
    "df.CLASS.value_counts().plot(kind='pie', autopct='%1.0f%%', colors=colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "136cfd5f-fa52-4d30-af3b-5f65cc6f5261",
   "metadata": {},
   "source": [
    "#### Remarks\n",
    "We can see that most tweets were either negative (46%) or neutral (44%). Only about 10% of tweets were positive."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48ac09a2-4c0f-4e63-81c3-d91cbb659a1e",
   "metadata": {},
   "source": [
    "#### 4.1.2. Exploring CLASS by inflow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a2e482f3-3c08-4878-8479-5ff60ce8f0c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Syrians       24627\n",
       "Ukrainians    13984\n",
       "Name: inflow, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the unique value counts of inflow (absolute frequency)\n",
    "\n",
    "df['inflow'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "41551d9c-a364-4a79-82d7-cf88b848f45a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Syrians       0.637823\n",
       "Ukrainians    0.362177\n",
       "Name: inflow, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the relative frequencies of inflow\n",
    "\n",
    "df['inflow'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f8ba9d39-f752-48d3-9232-3fadd0d5d098",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "inflow      CLASS   \n",
       "Syrians     NEUTRAL     0.545986\n",
       "            NEGATIVE    0.366427\n",
       "            POSITIVE    0.087587\n",
       "Ukrainians  NEGATIVE    0.620566\n",
       "            NEUTRAL     0.260154\n",
       "            POSITIVE    0.119279\n",
       "Name: CLASS, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding the proportion of classes by inflow\n",
    "\n",
    "byinflow = df.groupby('inflow').CLASS.value_counts(normalize=True)\n",
    "byinflow"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11ca7103-9a9f-4637-a2a2-a08634ac9753",
   "metadata": {},
   "source": [
    "#### Remarks\n",
    "For the Syrian inflow, most (~ 54.6%) tweets were neutral. Slightly over a third (~ 36.6%) were negative, and only about 8.8% were positive.\n",
    "\n",
    "For the Ukrainian inflow, most (~ 62.1%) were negative. Slightly under a third (~ 26%) were neutral, and only about 11.9% were positive."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8882384c-230f-48e6-af65-6158c28b50f1",
   "metadata": {},
   "source": [
    "### 4.3 Exploring by date\n",
    "#### 4.3.1 Moving Average Sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "93c73500-c975-446a-b38c-64661db1b52e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2015-09    4909\n",
       "2022-03    3116\n",
       "2015-10    2823\n",
       "2015-11    2603\n",
       "2016-01    2403\n",
       "2016-02    2217\n",
       "2015-08    2188\n",
       "2015-12    1726\n",
       "2016-03    1712\n",
       "2022-04    1469\n",
       "2022-09    1406\n",
       "2022-05    1338\n",
       "2022-10    1135\n",
       "2022-06    1022\n",
       "2022-08    1002\n",
       "2022-07    1001\n",
       "2022-02     922\n",
       "2015-07     876\n",
       "2016-04     734\n",
       "2015-04     521\n",
       "2015-06     520\n",
       "2015-05     405\n",
       "2015-01     339\n",
       "2021-11     331\n",
       "2021-08     314\n",
       "2015-02     287\n",
       "2015-03     245\n",
       "2022-01     239\n",
       "2021-10     192\n",
       "2021-09     165\n",
       "2021-12     157\n",
       "2021-07     128\n",
       "2014-12     119\n",
       "2021-06      47\n",
       "Name: year-month, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['year-month'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "b0483367-6ea6-410f-89b8-4260d9fc8b0a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# making new date_time variable\n",
    "\n",
    "df[\"date_time\"] = pd.to_datetime(df[\"created_at\"], format = '%Y-%m-%d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ff6ffb3d-3c51-4b2d-ad53-b7017822744b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# getting rid of timezone\n",
    "\n",
    "df['date_time'] = df['date_time'].apply(lambda x: x.replace(tzinfo=None))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4d9fb614-dc8c-4962-82bd-dba8ca0aa355",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>date</th>\n",
       "      <th>inflow</th>\n",
       "      <th>new text</th>\n",
       "      <th>scores</th>\n",
       "      <th>Negative</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "      <th>date_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>['0.0679844', '0.88380396', '0.0482117', '']</td>\n",
       "      <td>0.067984</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 22:55:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>['0.3660313', '0.5833739', '0.05059479']</td>\n",
       "      <td>0.366031</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:27:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>['0.30297568', '0.4430802', '0.25394407']</td>\n",
       "      <td>0.302976</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:18:58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>['0.02260253', '0.07604674', '0.9013506', '']</td>\n",
       "      <td>0.022603</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-04-20 18:25:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2016-04-20 00:00:00+00:00</td>\n",
       "      <td>Syrians</td>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>['0.6982864', '0.27545568', '0.02625783']</td>\n",
       "      <td>0.698286</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2016-04-20 16:27:28</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>38606</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-06-28 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>For day 1 of week 2, @user discusses the Socia...</td>\n",
       "      <td>['0.03550766', '0.82208896', '0.14240335']</td>\n",
       "      <td>0.035508</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-28 14:43:31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38607</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-06-27 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>@user Europe is racist, but in a different way...</td>\n",
       "      <td>['0.8931986', '0.09380516', '0.01299612']</td>\n",
       "      <td>0.893199</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-27 13:03:29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38608</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-06-27 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>['0.15655968', '0.5201203', '0.32332003']</td>\n",
       "      <td>0.156560</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-27 08:37:21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38609</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-06-26 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>@user I must have missed the great flight of t...</td>\n",
       "      <td>['0.91320664', '0.06375945', '0.02303378']</td>\n",
       "      <td>0.913207</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-26 08:03:22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38610</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2021-06-25 00:00:00+00:00</td>\n",
       "      <td>Ukrainians</td>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>['0.92373', '0.06414817', '0.01212183']</td>\n",
       "      <td>0.923730</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-25 21:02:17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "38606  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "38607  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "38609  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "38610  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "38606  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "38607  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "38608  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "38609  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "38610  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "38606                           0.0                         1.0   \n",
       "38607                           0.0                         1.0   \n",
       "38608                           0.0                         1.0   \n",
       "38609                           0.0                         0.0   \n",
       "38610                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "38606                        2.0                         0.0   \n",
       "38607                        6.0                         0.0   \n",
       "38608                       17.0                         0.0   \n",
       "38609                        1.0                         0.0   \n",
       "38610                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ...                       date  \\\n",
       "0                                  0.0  ...  2016-04-20 00:00:00+00:00   \n",
       "1                                  0.0  ...  2016-04-20 00:00:00+00:00   \n",
       "2                                  0.0  ...  2016-04-20 00:00:00+00:00   \n",
       "3                                  0.0  ...  2016-04-20 00:00:00+00:00   \n",
       "4                                  0.0  ...  2016-04-20 00:00:00+00:00   \n",
       "...                                ...  ...                        ...   \n",
       "38606                              0.0  ...  2021-06-28 00:00:00+00:00   \n",
       "38607                              0.0  ...  2021-06-27 00:00:00+00:00   \n",
       "38608                              0.0  ...  2021-06-27 00:00:00+00:00   \n",
       "38609                              0.0  ...  2021-06-26 00:00:00+00:00   \n",
       "38610                              0.0  ...  2021-06-25 00:00:00+00:00   \n",
       "\n",
       "           inflow                                           new text  \\\n",
       "0         Syrians  Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...   \n",
       "1         Syrians  Habe schon lang nicht gehört, daß Flüchtling G...   \n",
       "2         Syrians  \"Es kommen kaum noch Flüchtlinge nach Griechen...   \n",
       "3         Syrians  Unsere 1. Kochshow für #Flüchtlinge. Super spi...   \n",
       "4         Syrians  500 tote #Flüchtlinge im #Mittelmeer – Tragödi...   \n",
       "...           ...                                                ...   \n",
       "38606  Ukrainians  For day 1 of week 2, @user discusses the Socia...   \n",
       "38607  Ukrainians  @user Europe is racist, but in a different way...   \n",
       "38608  Ukrainians  A labour of love, inspired by Middle-earth.\\n\\...   \n",
       "38609  Ukrainians  @user I must have missed the great flight of t...   \n",
       "38610  Ukrainians  #Wuerzburg today an African immigrant or refug...   \n",
       "\n",
       "                                              scores  Negative   Neutral  \\\n",
       "0       ['0.0679844', '0.88380396', '0.0482117', '']  0.067984  0.883804   \n",
       "1           ['0.3660313', '0.5833739', '0.05059479']  0.366031  0.583374   \n",
       "2          ['0.30297568', '0.4430802', '0.25394407']  0.302976  0.443080   \n",
       "3      ['0.02260253', '0.07604674', '0.9013506', '']  0.022603  0.076047   \n",
       "4          ['0.6982864', '0.27545568', '0.02625783']  0.698286  0.275456   \n",
       "...                                              ...       ...       ...   \n",
       "38606     ['0.03550766', '0.82208896', '0.14240335']  0.035508  0.822089   \n",
       "38607      ['0.8931986', '0.09380516', '0.01299612']  0.893199  0.093805   \n",
       "38608      ['0.15655968', '0.5201203', '0.32332003']  0.156560  0.520120   \n",
       "38609     ['0.91320664', '0.06375945', '0.02303378']  0.913207  0.063759   \n",
       "38610        ['0.92373', '0.06414817', '0.01212183']  0.923730  0.064148   \n",
       "\n",
       "       Positive     CLASS num_CLASS           date_time  \n",
       "0      0.048212   NEUTRAL         0 2016-04-20 22:55:08  \n",
       "1      0.050595   NEUTRAL         0 2016-04-20 21:27:37  \n",
       "2      0.253944   NEUTRAL         0 2016-04-20 21:18:58  \n",
       "3      0.901351  POSITIVE         1 2016-04-20 18:25:11  \n",
       "4      0.026258  NEGATIVE        -1 2016-04-20 16:27:28  \n",
       "...         ...       ...       ...                 ...  \n",
       "38606  0.142403   NEUTRAL         0 2021-06-28 14:43:31  \n",
       "38607  0.012996  NEGATIVE        -1 2021-06-27 13:03:29  \n",
       "38608  0.323320   NEUTRAL         0 2021-06-27 08:37:21  \n",
       "38609  0.023034  NEGATIVE        -1 2021-06-26 08:03:22  \n",
       "38610  0.012122  NEGATIVE        -1 2021-06-25 21:02:17  \n",
       "\n",
       "[38611 rows x 29 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "bc956895-0bfd-46c6-bec8-70c145da86f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating Moving Averages of num_CLASS\n",
    "\n",
    "syrian = df[df['inflow']=='Syrians'][['date_time', 'num_CLASS']]\n",
    "syrian = syrian.sort_values(by='date_time', ascending=True)\n",
    "syrian['Moving Average Sentiment'] = syrian.num_CLASS.rolling(20, min_periods=3).mean()\n",
    "\n",
    "ukrainian = df[df['inflow']=='Ukrainians'][['date_time', 'num_CLASS']]\n",
    "ukrainian = ukrainian.sort_values(by='date_time', ascending=True)\n",
    "ukrainian['Moving Average Sentiment'] = ukrainian.num_CLASS.rolling(20, min_periods=3).mean()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6a38b9bf-3575-4d8e-8e85-ea815bc4405a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "year-month\n",
       "2014-12   -0.319328\n",
       "2015-01   -0.380531\n",
       "2015-02   -0.257840\n",
       "2015-03   -0.212245\n",
       "2015-04   -0.303263\n",
       "2015-05   -0.246914\n",
       "2015-06   -0.251923\n",
       "2015-07   -0.295662\n",
       "2015-08   -0.318556\n",
       "2015-09   -0.228763\n",
       "2015-10   -0.263549\n",
       "2015-11   -0.292739\n",
       "2015-12   -0.209154\n",
       "2016-01   -0.348315\n",
       "2016-02   -0.321155\n",
       "2016-03   -0.276869\n",
       "2016-04   -0.294278\n",
       "2021-06   -0.617021\n",
       "2021-07   -0.546875\n",
       "2021-08   -0.608280\n",
       "2021-09   -0.442424\n",
       "2021-10   -0.500000\n",
       "2021-11   -0.583082\n",
       "2021-12   -0.394904\n",
       "2022-01   -0.615063\n",
       "2022-02   -0.462039\n",
       "2022-03   -0.401797\n",
       "2022-04   -0.504425\n",
       "2022-05   -0.491779\n",
       "2022-06   -0.495108\n",
       "2022-07   -0.527473\n",
       "2022-08   -0.556886\n",
       "2022-09   -0.557610\n",
       "2022-10   -0.613216\n",
       "Name: num_CLASS, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding monthly average sentiment\n",
    "\n",
    "df.groupby('year-month').num_CLASS.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "15b3aef2-e8e7-4371-a502-5f820ab53589",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.style.use('tableau-colorblind10')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0b5e621e-d214-43c1-8a25-f50297932c26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1300x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting Moving Averages for each inflow using colorblind-friendly colors & Times New Roman\n",
    "\n",
    "csfont = {'fontname':'Times New Roman'}\n",
    "\n",
    "fix, axes = plt.subplots(2,1, figsize=(13,10))\n",
    "custom_ylim = (-1,1)\n",
    "\n",
    "plt.setp(axes, ylim=custom_ylim)\n",
    "\n",
    "axes[0].plot(syrian['date_time'], syrian['Moving Average Sentiment'], color='#FFBC79')\n",
    "axes[0].set_title(\"\\n\".join(['Syrian Inflow']), **csfont)\n",
    "axes[0].axhline(0, color='lightgrey', ls='--')\n",
    "\n",
    "# vertical line (center time point)\n",
    "ax_vline = axes[0].axvline(datetime(2015, 8, 21), color = 'red', ls='dotted', label='Halt of Dublin Regulation')\n",
    "leg = axes[0].legend(handles=[ax_vline])\n",
    "\n",
    "# set font for the legend text\n",
    "for text in leg.get_texts():\n",
    "    text.set_fontname('Times New Roman')\n",
    "    \n",
    "# set font for tick labels\n",
    "for tick in axes[0].get_xticklabels() + axes[0].get_yticklabels():\n",
    "    tick.set_fontname('Times New Roman')\n",
    "    \n",
    "# text annotations for average monthly sentiment\n",
    "syrian_avg = syrian.groupby(pd.Grouper(key='date_time', freq='M'))['Moving Average Sentiment'].mean()\n",
    "for i, v in syrian_avg.items():\n",
    "    axes[0].annotate(f'{v:.2f}', xy=(i, v), xytext=(-10, -30), textcoords='offset points', fontsize=10, ha='center', va='top', **csfont)\n",
    "\n",
    "axes[1].plot(ukrainian['date_time'], ukrainian['Moving Average Sentiment'], color='#A2C8EC')\n",
    "axes[1].set_title(\"\\n\".join(['Ukrainian Inflow']), **csfont)\n",
    "axes[1].axhline(0, color='lightgrey', ls='--')\n",
    "\n",
    "#vertical line (center time point)\n",
    "ax_vline = axes[1].axvline(datetime(2022, 2, 24), color = 'red', ls='dotted', label=\"Russia's invasion of Ukraine\")\n",
    "leg = axes[1].legend(handles=[ax_vline])\n",
    "\n",
    "# set font for the legend text\n",
    "for text in leg.get_texts():\n",
    "    text.set_fontname('Times New Roman')\n",
    "    \n",
    "# set font for tick labels\n",
    "for tick in axes[1].get_xticklabels() + axes[1].get_yticklabels():\n",
    "    tick.set_fontname('Times New Roman')\n",
    "    \n",
    "# text annotations for average monthly sentiment\n",
    "ukrainian_avg = ukrainian.groupby(pd.Grouper(key='date_time', freq='M'))['Moving Average Sentiment'].mean()\n",
    "for i, v in ukrainian_avg.items():\n",
    "    axes[1].annotate(f'{v:.2f}', xy=(i, v), xytext=(-10, -30), textcoords='offset points', fontsize=10, ha='center', va='top', **csfont)\n",
    "\n",
    "plt.savefig('10-Tweet_Moving-Avg_colorblind.png', bbox_inches='tight' , dpi=200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c9e1697-9867-4aa3-b3a3-36356fdfe36a",
   "metadata": {},
   "source": [
    "#### 4.3.2. Before or during inflow?\n",
    "I want to create a variable indicating whether its during the inflow (after center point) or before\n",
    "* during syrian inflow: from 2015-08-21 onwards \n",
    "* before syrian inflow: before 2015-08-21\n",
    "\n",
    "* during ukrainian inflow: from 2022-02-24 onwards\n",
    "* before ukrainian inflow: before 2022-02-24 (but after April 21, 2016)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "f53e75bb-442b-4c29-b4ac-8422ed03fe11",
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating Syrian and Ukrainian datasets\n",
    "\n",
    "df_syr = df[df.inflow=='Syrians'].copy()\n",
    "df_uk = df[df.inflow=='Ukrainians'].copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce43f845-5b62-40f5-9d35-13618bb752fb",
   "metadata": {},
   "source": [
    "#### *Before inflow*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "347bca6f-3fca-4687-afa0-339ea3e8ba94",
   "metadata": {},
   "outputs": [],
   "source": [
    "# assigning True to all tweets before syrian inflow\n",
    "\n",
    "df['before_syrian_inflow'] = df['date_time'] < '2015-08-21'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0d7c9216-82bb-42eb-9e16-c94b10d76055",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    34324\n",
       "True      4287\n",
       "Name: before_syrian_inflow, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['before_syrian_inflow'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "d13d7342-9f68-4492-8f2f-83814091a392",
   "metadata": {},
   "outputs": [],
   "source": [
    "# assigning True to all tweets before ukrainian inflow \n",
    "\n",
    "df['before_ukrainian_inflow'] = (df['date_time'] < '2022-02-24') & (df['inflow']=='Ukrainians')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "7c95a065-2f6a-4795-bcb7-26e0ba7c8199",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    36809\n",
       "True      1802\n",
       "Name: before_ukrainian_inflow, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['before_ukrainian_inflow'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "29843bac-67b1-4184-bc85-a5c8b0701b66",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>new text</th>\n",
       "      <th>scores</th>\n",
       "      <th>Negative</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "      <th>date_time</th>\n",
       "      <th>before_syrian_inflow</th>\n",
       "      <th>before_ukrainian_inflow</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>['0.0679844', '0.88380396', '0.0482117', '']</td>\n",
       "      <td>0.067984</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 22:55:08</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>['0.3660313', '0.5833739', '0.05059479']</td>\n",
       "      <td>0.366031</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:27:37</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>['0.30297568', '0.4430802', '0.25394407']</td>\n",
       "      <td>0.302976</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:18:58</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>['0.02260253', '0.07604674', '0.9013506', '']</td>\n",
       "      <td>0.022603</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-04-20 18:25:11</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>['0.6982864', '0.27545568', '0.02625783']</td>\n",
       "      <td>0.698286</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2016-04-20 16:27:28</td>\n",
       "      <td>False</td>\n",
       "      <td>False</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>38606</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>For day 1 of week 2, @user discusses the Socia...</td>\n",
       "      <td>['0.03550766', '0.82208896', '0.14240335']</td>\n",
       "      <td>0.035508</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-28 14:43:31</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38607</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>@user Europe is racist, but in a different way...</td>\n",
       "      <td>['0.8931986', '0.09380516', '0.01299612']</td>\n",
       "      <td>0.893199</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-27 13:03:29</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38608</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>['0.15655968', '0.5201203', '0.32332003']</td>\n",
       "      <td>0.156560</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-27 08:37:21</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38609</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>@user I must have missed the great flight of t...</td>\n",
       "      <td>['0.91320664', '0.06375945', '0.02303378']</td>\n",
       "      <td>0.913207</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-26 08:03:22</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38610</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>['0.92373', '0.06414817', '0.01212183']</td>\n",
       "      <td>0.923730</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-25 21:02:17</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "38606  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "38607  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "38609  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "38610  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "38606  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "38607  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "38608  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "38609  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "38610  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "38606                           0.0                         1.0   \n",
       "38607                           0.0                         1.0   \n",
       "38608                           0.0                         1.0   \n",
       "38609                           0.0                         0.0   \n",
       "38610                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "38606                        2.0                         0.0   \n",
       "38607                        6.0                         0.0   \n",
       "38608                       17.0                         0.0   \n",
       "38609                        1.0                         0.0   \n",
       "38610                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ...  \\\n",
       "0                                  0.0  ...   \n",
       "1                                  0.0  ...   \n",
       "2                                  0.0  ...   \n",
       "3                                  0.0  ...   \n",
       "4                                  0.0  ...   \n",
       "...                                ...  ...   \n",
       "38606                              0.0  ...   \n",
       "38607                              0.0  ...   \n",
       "38608                              0.0  ...   \n",
       "38609                              0.0  ...   \n",
       "38610                              0.0  ...   \n",
       "\n",
       "                                                new text  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...   \n",
       "...                                                  ...   \n",
       "38606  For day 1 of week 2, @user discusses the Socia...   \n",
       "38607  @user Europe is racist, but in a different way...   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...   \n",
       "38609  @user I must have missed the great flight of t...   \n",
       "38610  #Wuerzburg today an African immigrant or refug...   \n",
       "\n",
       "                                              scores  Negative   Neutral  \\\n",
       "0       ['0.0679844', '0.88380396', '0.0482117', '']  0.067984  0.883804   \n",
       "1           ['0.3660313', '0.5833739', '0.05059479']  0.366031  0.583374   \n",
       "2          ['0.30297568', '0.4430802', '0.25394407']  0.302976  0.443080   \n",
       "3      ['0.02260253', '0.07604674', '0.9013506', '']  0.022603  0.076047   \n",
       "4          ['0.6982864', '0.27545568', '0.02625783']  0.698286  0.275456   \n",
       "...                                              ...       ...       ...   \n",
       "38606     ['0.03550766', '0.82208896', '0.14240335']  0.035508  0.822089   \n",
       "38607      ['0.8931986', '0.09380516', '0.01299612']  0.893199  0.093805   \n",
       "38608      ['0.15655968', '0.5201203', '0.32332003']  0.156560  0.520120   \n",
       "38609     ['0.91320664', '0.06375945', '0.02303378']  0.913207  0.063759   \n",
       "38610        ['0.92373', '0.06414817', '0.01212183']  0.923730  0.064148   \n",
       "\n",
       "       Positive     CLASS num_CLASS           date_time before_syrian_inflow  \\\n",
       "0      0.048212   NEUTRAL         0 2016-04-20 22:55:08                False   \n",
       "1      0.050595   NEUTRAL         0 2016-04-20 21:27:37                False   \n",
       "2      0.253944   NEUTRAL         0 2016-04-20 21:18:58                False   \n",
       "3      0.901351  POSITIVE         1 2016-04-20 18:25:11                False   \n",
       "4      0.026258  NEGATIVE        -1 2016-04-20 16:27:28                False   \n",
       "...         ...       ...       ...                 ...                  ...   \n",
       "38606  0.142403   NEUTRAL         0 2021-06-28 14:43:31                False   \n",
       "38607  0.012996  NEGATIVE        -1 2021-06-27 13:03:29                False   \n",
       "38608  0.323320   NEUTRAL         0 2021-06-27 08:37:21                False   \n",
       "38609  0.023034  NEGATIVE        -1 2021-06-26 08:03:22                False   \n",
       "38610  0.012122  NEGATIVE        -1 2021-06-25 21:02:17                False   \n",
       "\n",
       "      before_ukrainian_inflow  \n",
       "0                       False  \n",
       "1                       False  \n",
       "2                       False  \n",
       "3                       False  \n",
       "4                       False  \n",
       "...                       ...  \n",
       "38606                    True  \n",
       "38607                    True  \n",
       "38608                    True  \n",
       "38609                    True  \n",
       "38610                    True  \n",
       "\n",
       "[38611 rows x 31 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "214c6d0d-514c-4fbd-a0fc-56302ea3b012",
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating 'before_inflow'\n",
    "\n",
    "df['before_inflow'] = (df['before_syrian_inflow']==True) | (df['before_ukrainian_inflow']==True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "aec984b8-9958-4ea3-84dc-0bba5a31aef5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    32522\n",
       "True      6089\n",
       "Name: before_inflow, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['before_inflow'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3ca0d2ec-4f39-44ae-8125-fa987c839f67",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>scores</th>\n",
       "      <th>Negative</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "      <th>date_time</th>\n",
       "      <th>before_syrian_inflow</th>\n",
       "      <th>before_ukrainian_inflow</th>\n",
       "      <th>before_inflow</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.0679844', '0.88380396', '0.0482117', '']</td>\n",
       "      <td>0.067984</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 22:55:08</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.3660313', '0.5833739', '0.05059479']</td>\n",
       "      <td>0.366031</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:27:37</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.30297568', '0.4430802', '0.25394407']</td>\n",
       "      <td>0.302976</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:18:58</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.02260253', '0.07604674', '0.9013506', '']</td>\n",
       "      <td>0.022603</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-04-20 18:25:11</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.6982864', '0.27545568', '0.02625783']</td>\n",
       "      <td>0.698286</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2016-04-20 16:27:28</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>38606</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.03550766', '0.82208896', '0.14240335']</td>\n",
       "      <td>0.035508</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-28 14:43:31</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38607</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.8931986', '0.09380516', '0.01299612']</td>\n",
       "      <td>0.893199</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-27 13:03:29</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38608</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.15655968', '0.5201203', '0.32332003']</td>\n",
       "      <td>0.156560</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-27 08:37:21</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38609</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.91320664', '0.06375945', '0.02303378']</td>\n",
       "      <td>0.913207</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-26 08:03:22</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38610</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>['0.92373', '0.06414817', '0.01212183']</td>\n",
       "      <td>0.923730</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-25 21:02:17</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "38606  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "38607  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "38609  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "38610  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "38606  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "38607  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "38608  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "38609  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "38610  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "38606                           0.0                         1.0   \n",
       "38607                           0.0                         1.0   \n",
       "38608                           0.0                         1.0   \n",
       "38609                           0.0                         0.0   \n",
       "38610                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "38606                        2.0                         0.0   \n",
       "38607                        6.0                         0.0   \n",
       "38608                       17.0                         0.0   \n",
       "38609                        1.0                         0.0   \n",
       "38610                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ...  \\\n",
       "0                                  0.0  ...   \n",
       "1                                  0.0  ...   \n",
       "2                                  0.0  ...   \n",
       "3                                  0.0  ...   \n",
       "4                                  0.0  ...   \n",
       "...                                ...  ...   \n",
       "38606                              0.0  ...   \n",
       "38607                              0.0  ...   \n",
       "38608                              0.0  ...   \n",
       "38609                              0.0  ...   \n",
       "38610                              0.0  ...   \n",
       "\n",
       "                                              scores  Negative   Neutral  \\\n",
       "0       ['0.0679844', '0.88380396', '0.0482117', '']  0.067984  0.883804   \n",
       "1           ['0.3660313', '0.5833739', '0.05059479']  0.366031  0.583374   \n",
       "2          ['0.30297568', '0.4430802', '0.25394407']  0.302976  0.443080   \n",
       "3      ['0.02260253', '0.07604674', '0.9013506', '']  0.022603  0.076047   \n",
       "4          ['0.6982864', '0.27545568', '0.02625783']  0.698286  0.275456   \n",
       "...                                              ...       ...       ...   \n",
       "38606     ['0.03550766', '0.82208896', '0.14240335']  0.035508  0.822089   \n",
       "38607      ['0.8931986', '0.09380516', '0.01299612']  0.893199  0.093805   \n",
       "38608      ['0.15655968', '0.5201203', '0.32332003']  0.156560  0.520120   \n",
       "38609     ['0.91320664', '0.06375945', '0.02303378']  0.913207  0.063759   \n",
       "38610        ['0.92373', '0.06414817', '0.01212183']  0.923730  0.064148   \n",
       "\n",
       "       Positive     CLASS  num_CLASS           date_time before_syrian_inflow  \\\n",
       "0      0.048212   NEUTRAL          0 2016-04-20 22:55:08                False   \n",
       "1      0.050595   NEUTRAL          0 2016-04-20 21:27:37                False   \n",
       "2      0.253944   NEUTRAL          0 2016-04-20 21:18:58                False   \n",
       "3      0.901351  POSITIVE          1 2016-04-20 18:25:11                False   \n",
       "4      0.026258  NEGATIVE         -1 2016-04-20 16:27:28                False   \n",
       "...         ...       ...        ...                 ...                  ...   \n",
       "38606  0.142403   NEUTRAL          0 2021-06-28 14:43:31                False   \n",
       "38607  0.012996  NEGATIVE         -1 2021-06-27 13:03:29                False   \n",
       "38608  0.323320   NEUTRAL          0 2021-06-27 08:37:21                False   \n",
       "38609  0.023034  NEGATIVE         -1 2021-06-26 08:03:22                False   \n",
       "38610  0.012122  NEGATIVE         -1 2021-06-25 21:02:17                False   \n",
       "\n",
       "      before_ukrainian_inflow before_inflow  \n",
       "0                       False         False  \n",
       "1                       False         False  \n",
       "2                       False         False  \n",
       "3                       False         False  \n",
       "4                       False         False  \n",
       "...                       ...           ...  \n",
       "38606                    True          True  \n",
       "38607                    True          True  \n",
       "38608                    True          True  \n",
       "38609                    True          True  \n",
       "38610                    True          True  \n",
       "\n",
       "[38611 rows x 32 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "61cf2e01-8140-473b-afa3-18f12e468597",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    0.842299\n",
       "True     0.157701\n",
       "Name: before_inflow, dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['before_inflow'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "011e63dd-6dd7-48c1-9425-872e9e802678",
   "metadata": {},
   "source": [
    "#### Remarks\n",
    "We see that approx. 84 percent of all tweets were published during the inflow rather than before."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe1fd15e-7837-491f-afc0-fc80af1ecf52",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### *During inflow*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "a941f6b4-a3a0-4219-a4d0-b0d2c04cbd05",
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating 'during_inflow'\n",
    "\n",
    "df['during_inflow'] = df['before_inflow']==False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "765ac43c-5081-46ee-95ec-92f33ad8fe92",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>Negative</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "      <th>date_time</th>\n",
       "      <th>before_syrian_inflow</th>\n",
       "      <th>before_ukrainian_inflow</th>\n",
       "      <th>before_inflow</th>\n",
       "      <th>during_inflow</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.067984</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 22:55:08</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.366031</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:27:37</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.302976</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:18:58</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.022603</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-04-20 18:25:11</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.698286</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2016-04-20 16:27:28</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>38606</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.035508</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-28 14:43:31</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38607</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.893199</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-27 13:03:29</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38608</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.156560</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-27 08:37:21</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38609</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.913207</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-26 08:03:22</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38610</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.923730</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-25 21:02:17</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "38606  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "38607  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "38609  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "38610  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "38606  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "38607  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "38608  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "38609  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "38610  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "38606                           0.0                         1.0   \n",
       "38607                           0.0                         1.0   \n",
       "38608                           0.0                         1.0   \n",
       "38609                           0.0                         0.0   \n",
       "38610                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "38606                        2.0                         0.0   \n",
       "38607                        6.0                         0.0   \n",
       "38608                       17.0                         0.0   \n",
       "38609                        1.0                         0.0   \n",
       "38610                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ...  Negative   Neutral  Positive  \\\n",
       "0                                  0.0  ...  0.067984  0.883804  0.048212   \n",
       "1                                  0.0  ...  0.366031  0.583374  0.050595   \n",
       "2                                  0.0  ...  0.302976  0.443080  0.253944   \n",
       "3                                  0.0  ...  0.022603  0.076047  0.901351   \n",
       "4                                  0.0  ...  0.698286  0.275456  0.026258   \n",
       "...                                ...  ...       ...       ...       ...   \n",
       "38606                              0.0  ...  0.035508  0.822089  0.142403   \n",
       "38607                              0.0  ...  0.893199  0.093805  0.012996   \n",
       "38608                              0.0  ...  0.156560  0.520120  0.323320   \n",
       "38609                              0.0  ...  0.913207  0.063759  0.023034   \n",
       "38610                              0.0  ...  0.923730  0.064148  0.012122   \n",
       "\n",
       "          CLASS  num_CLASS           date_time before_syrian_inflow  \\\n",
       "0       NEUTRAL          0 2016-04-20 22:55:08                False   \n",
       "1       NEUTRAL          0 2016-04-20 21:27:37                False   \n",
       "2       NEUTRAL          0 2016-04-20 21:18:58                False   \n",
       "3      POSITIVE          1 2016-04-20 18:25:11                False   \n",
       "4      NEGATIVE         -1 2016-04-20 16:27:28                False   \n",
       "...         ...        ...                 ...                  ...   \n",
       "38606   NEUTRAL          0 2021-06-28 14:43:31                False   \n",
       "38607  NEGATIVE         -1 2021-06-27 13:03:29                False   \n",
       "38608   NEUTRAL          0 2021-06-27 08:37:21                False   \n",
       "38609  NEGATIVE         -1 2021-06-26 08:03:22                False   \n",
       "38610  NEGATIVE         -1 2021-06-25 21:02:17                False   \n",
       "\n",
       "      before_ukrainian_inflow before_inflow during_inflow  \n",
       "0                       False         False          True  \n",
       "1                       False         False          True  \n",
       "2                       False         False          True  \n",
       "3                       False         False          True  \n",
       "4                       False         False          True  \n",
       "...                       ...           ...           ...  \n",
       "38606                    True          True         False  \n",
       "38607                    True          True         False  \n",
       "38608                    True          True         False  \n",
       "38609                    True          True         False  \n",
       "38610                    True          True         False  \n",
       "\n",
       "[38611 rows x 33 columns]"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "757a7377-f431-46de-bd5d-51561f196907",
   "metadata": {},
   "outputs": [],
   "source": [
    "# re-establishing syrian & ukrainian dataset (to include to 'before_inflow' variable)\n",
    "\n",
    "df_syr = df[df.inflow == 'Syrians'].copy()\n",
    "\n",
    "df_uk = df[df.inflow == 'Ukrainians'].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "9adbc292-c93c-4fc7-9642-df0b19351ad3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "before_inflow  CLASS   \n",
       "False          NEUTRAL     0.546313\n",
       "               NEGATIVE    0.365093\n",
       "               POSITIVE    0.088594\n",
       "True           NEUTRAL     0.544437\n",
       "               NEGATIVE    0.372755\n",
       "               POSITIVE    0.082808\n",
       "Name: CLASS, dtype: float64"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding the proportion of classes by before or during syrian inflow\n",
    "\n",
    "bybord = df_syr.groupby('before_inflow').CLASS.value_counts(normalize=True)\n",
    "bybord"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "b44889a8-579b-482f-ba69-a35858969c50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "before_inflow  CLASS   \n",
       "False          NEGATIVE    0.618371\n",
       "               NEUTRAL     0.258168\n",
       "               POSITIVE    0.123461\n",
       "True           NEGATIVE    0.635405\n",
       "               NEUTRAL     0.273585\n",
       "               POSITIVE    0.091010\n",
       "Name: CLASS, dtype: float64"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding the proportion of classes by before or during ukrainian inflow\n",
    "\n",
    "bybord = df_uk.groupby('before_inflow').CLASS.value_counts(normalize=True)\n",
    "bybord"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4e7084f-18d0-44b9-bc1c-fa39f6063c12",
   "metadata": {},
   "source": [
    "### 4.4 Exploring num_CLASS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "e31e803f-2159-4778-a1b6-a4d109d9c5d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    38611.000000\n",
       "mean        -0.359405\n",
       "std          0.654503\n",
       "min         -1.000000\n",
       "25%         -1.000000\n",
       "50%          0.000000\n",
       "75%          0.000000\n",
       "max          1.000000\n",
       "Name: num_CLASS, dtype: float64"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "2b9d4bee-c0fc-46d0-beb0-1234159f3ede",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    24627.000000\n",
       "mean        -0.278840\n",
       "std          0.613414\n",
       "min         -1.000000\n",
       "25%         -1.000000\n",
       "50%          0.000000\n",
       "75%          0.000000\n",
       "max          1.000000\n",
       "Name: num_CLASS, dtype: float64"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining num_CLASS for Syrian dataset\n",
    "\n",
    "df_syr.num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "44b0d405-260a-4e63-a81b-bc0324dea44a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    13984.000000\n",
       "mean        -0.501287\n",
       "std          0.698993\n",
       "min         -1.000000\n",
       "25%         -1.000000\n",
       "50%         -1.000000\n",
       "75%          0.000000\n",
       "max          1.000000\n",
       "Name: num_CLASS, dtype: float64"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining num_CLASS for Ukrainian dataset\n",
    "\n",
    "df_uk.num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "558e8ca9-1438-4f67-83ea-b4e7668d315a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating a dataset for before inflow and during inflow\n",
    "\n",
    "df_before = df[df.during_inflow == False].copy()\n",
    "df_during = df[df.during_inflow == True].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "fef7cbd6-8434-495e-9837-b6f9dd7da795",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    6089.000000\n",
       "mean       -0.365249\n",
       "std         0.634334\n",
       "min        -1.000000\n",
       "25%        -1.000000\n",
       "50%         0.000000\n",
       "75%         0.000000\n",
       "max         1.000000\n",
       "Name: num_CLASS, dtype: float64"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_before.num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "031a230d-b84d-4cc9-84a8-93c91b417920",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    32522.000000\n",
       "mean        -0.358311\n",
       "std          0.658214\n",
       "min         -1.000000\n",
       "25%         -1.000000\n",
       "50%          0.000000\n",
       "75%          0.000000\n",
       "max          1.000000\n",
       "Name: num_CLASS, dtype: float64"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_during.num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "1e4ceb7f-0685-4214-8846-cf03eb041e94",
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inflow</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Syrians</th>\n",
       "      <td>4287.0</td>\n",
       "      <td>-0.289946</td>\n",
       "      <td>0.609575</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ukrainians</th>\n",
       "      <td>1802.0</td>\n",
       "      <td>-0.544395</td>\n",
       "      <td>0.655963</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             count      mean       std  min  25%  50%  75%  max\n",
       "inflow                                                         \n",
       "Syrians     4287.0 -0.289946  0.609575 -1.0 -1.0  0.0  0.0  1.0\n",
       "Ukrainians  1802.0 -0.544395  0.655963 -1.0 -1.0 -1.0  0.0  1.0"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_before.groupby('inflow').num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "49031a76-9035-4086-a827-3e2cf383efed",
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inflow</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Syrians</th>\n",
       "      <td>20340.0</td>\n",
       "      <td>-0.276500</td>\n",
       "      <td>0.614210</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ukrainians</th>\n",
       "      <td>12182.0</td>\n",
       "      <td>-0.494911</td>\n",
       "      <td>0.704937</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              count      mean       std  min  25%  50%  75%  max\n",
       "inflow                                                          \n",
       "Syrians     20340.0 -0.276500  0.614210 -1.0 -1.0  0.0  0.0  1.0\n",
       "Ukrainians  12182.0 -0.494911  0.704937 -1.0 -1.0 -1.0  0.0  1.0"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_during.groupby('inflow').num_CLASS.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "201af09b-a715-4637-b394-4624fe04eb83",
   "metadata": {},
   "source": [
    "#### Remarks\n",
    "\n",
    "Before the Syrian inflow, the average sentiment of all tweets was about -0.29. During the inflow, the average sentiment became every so slightly less negative (-0.28).\n",
    "\n",
    "Before the Ukrainian inflow, the average sentiment of all tweets was about -0.54 but rose to about -0.49 during the inflow."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0d3e870-1aac-4e83-bcbd-0c83d631dcee",
   "metadata": {},
   "source": [
    "## 5. Exploring with binary CLASS (NEGATIVE, POSITIVE)\n",
    "Now, CLASS will by dichotomized into a new variable, CLASS_bin:\n",
    "* NEGATIVE = 1\n",
    "* NOT_NEGATIVE (POSITIVE / NEUTRAL) = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "0ebae81c-3e2b-4c76-9cbf-909f67f847e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NEGATIVE    17702\n",
       "NEUTRAL     17084\n",
       "POSITIVE     3825\n",
       "Name: CLASS, dtype: int64"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.CLASS.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "fb76177a-21c2-4b2d-8783-9b8172e6ea72",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20909\n"
     ]
    }
   ],
   "source": [
    "print(17084 + 3825)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a840bfe8-8c78-407a-8da9-9359733a5490",
   "metadata": {},
   "source": [
    "When NEUTRAL and POSITIVE are recoded to be NOT_NEGATIVE, NOT_NEGATIVE should contain 20909 value counts "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "00e42ffc-d28d-4556-86e3-380153f26897",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['CLASS_bin'] = ''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "2895cff8-4927-456f-aa0a-d9b28de5f0cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# defining a function to derive CLASS_bin\n",
    "\n",
    "def calc_CLASS_bin(CLASS):\n",
    "    if CLASS == 'NEGATIVE':\n",
    "        CLASS_bin = 1\n",
    "    elif CLASS == 'NEUTRAL':\n",
    "        CLASS_bin = 0\n",
    "    elif CLASS == 'POSITIVE':\n",
    "        CLASS_bin = 0\n",
    "    return CLASS_bin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "fe6d2c1f-bea0-41c4-9e67-d7bfc68776ab",
   "metadata": {},
   "outputs": [],
   "source": [
    "# applying function\n",
    "\n",
    "df['CLASS_bin'] = df.apply(lambda x: calc_CLASS_bin(x['CLASS']), \n",
    "                        axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "1377e3f0-74cf-4bfd-a7f2-98501954d93b",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>author_id</th>\n",
       "      <th>created_at</th>\n",
       "      <th>lang</th>\n",
       "      <th>geo.place_id</th>\n",
       "      <th>public_metrics.retweet_count</th>\n",
       "      <th>public_metrics.reply_count</th>\n",
       "      <th>public_metrics.like_count</th>\n",
       "      <th>public_metrics.quote_count</th>\n",
       "      <th>public_metrics.impression_count</th>\n",
       "      <th>...</th>\n",
       "      <th>Neutral</th>\n",
       "      <th>Positive</th>\n",
       "      <th>CLASS</th>\n",
       "      <th>num_CLASS</th>\n",
       "      <th>date_time</th>\n",
       "      <th>before_syrian_inflow</th>\n",
       "      <th>before_ukrainian_inflow</th>\n",
       "      <th>before_inflow</th>\n",
       "      <th>during_inflow</th>\n",
       "      <th>CLASS_bin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...</td>\n",
       "      <td>4.122038e+09</td>\n",
       "      <td>2016-04-20 22:55:08+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>06d9a7c249c59bcd</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.883804</td>\n",
       "      <td>0.048212</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 22:55:08</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Habe schon lang nicht gehört, daß Flüchtling G...</td>\n",
       "      <td>1.179544e+09</td>\n",
       "      <td>2016-04-20 21:27:37+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e99b714fe65be4fb</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.583374</td>\n",
       "      <td>0.050595</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:27:37</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Es kommen kaum noch Flüchtlinge nach Griechen...</td>\n",
       "      <td>2.246076e+08</td>\n",
       "      <td>2016-04-20 21:18:58+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>3078869807f9dd36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.443080</td>\n",
       "      <td>0.253944</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-04-20 21:18:58</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Unsere 1. Kochshow für #Flüchtlinge. Super spi...</td>\n",
       "      <td>2.480764e+09</td>\n",
       "      <td>2016-04-20 18:25:11+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>8abc99434d4f5d28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.076047</td>\n",
       "      <td>0.901351</td>\n",
       "      <td>POSITIVE</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-04-20 18:25:11</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>500 tote #Flüchtlinge im #Mittelmeer – Tragödi...</td>\n",
       "      <td>6.062653e+08</td>\n",
       "      <td>2016-04-20 16:27:28+00:00</td>\n",
       "      <td>de</td>\n",
       "      <td>e11a8b8e3771f9fa</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.275456</td>\n",
       "      <td>0.026258</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2016-04-20 16:27:28</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>1</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>38606</th>\n",
       "      <td>For day 1 of week 2, @AnnaMariaKonsta discusse...</td>\n",
       "      <td>1.104025e+08</td>\n",
       "      <td>2021-06-28 14:43:31+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>fcbb3c6e0a7eba22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.822089</td>\n",
       "      <td>0.142403</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-28 14:43:31</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38607</th>\n",
       "      <td>@ariadneconill Europe is racist, but in a diff...</td>\n",
       "      <td>2.521809e+09</td>\n",
       "      <td>2021-06-27 13:03:29+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>e385d4d639c6a423</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.093805</td>\n",
       "      <td>0.012996</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-27 13:03:29</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38608</th>\n",
       "      <td>A labour of love, inspired by Middle-earth.\\n\\...</td>\n",
       "      <td>5.633818e+08</td>\n",
       "      <td>2021-06-27 08:37:21+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>257640324f249a73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.520120</td>\n",
       "      <td>0.323320</td>\n",
       "      <td>NEUTRAL</td>\n",
       "      <td>0</td>\n",
       "      <td>2021-06-27 08:37:21</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38609</th>\n",
       "      <td>@simongerman600 I must have missed the great f...</td>\n",
       "      <td>2.591892e+09</td>\n",
       "      <td>2021-06-26 08:03:22+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>000b71538f35fe46</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.063759</td>\n",
       "      <td>0.023034</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-26 08:03:22</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38610</th>\n",
       "      <td>#Wuerzburg today an African immigrant or refug...</td>\n",
       "      <td>1.344017e+08</td>\n",
       "      <td>2021-06-25 21:02:17+00:00</td>\n",
       "      <td>en</td>\n",
       "      <td>08f94128283f7bc6</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.064148</td>\n",
       "      <td>0.012122</td>\n",
       "      <td>NEGATIVE</td>\n",
       "      <td>-1</td>\n",
       "      <td>2021-06-25 21:02:17</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>38611 rows × 34 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text     author_id  \\\n",
       "0      Syrisch-orthodoxer Bischof: \"Was im Nahen Oste...  4.122038e+09   \n",
       "1      Habe schon lang nicht gehört, daß Flüchtling G...  1.179544e+09   \n",
       "2      \"Es kommen kaum noch Flüchtlinge nach Griechen...  2.246076e+08   \n",
       "3      Unsere 1. Kochshow für #Flüchtlinge. Super spi...  2.480764e+09   \n",
       "4      500 tote #Flüchtlinge im #Mittelmeer – Tragödi...  6.062653e+08   \n",
       "...                                                  ...           ...   \n",
       "38606  For day 1 of week 2, @AnnaMariaKonsta discusse...  1.104025e+08   \n",
       "38607  @ariadneconill Europe is racist, but in a diff...  2.521809e+09   \n",
       "38608  A labour of love, inspired by Middle-earth.\\n\\...  5.633818e+08   \n",
       "38609  @simongerman600 I must have missed the great f...  2.591892e+09   \n",
       "38610  #Wuerzburg today an African immigrant or refug...  1.344017e+08   \n",
       "\n",
       "                      created_at lang      geo.place_id  \\\n",
       "0      2016-04-20 22:55:08+00:00   de  06d9a7c249c59bcd   \n",
       "1      2016-04-20 21:27:37+00:00   de  e99b714fe65be4fb   \n",
       "2      2016-04-20 21:18:58+00:00   de  3078869807f9dd36   \n",
       "3      2016-04-20 18:25:11+00:00   de  8abc99434d4f5d28   \n",
       "4      2016-04-20 16:27:28+00:00   de  e11a8b8e3771f9fa   \n",
       "...                          ...  ...               ...   \n",
       "38606  2021-06-28 14:43:31+00:00   en  fcbb3c6e0a7eba22   \n",
       "38607  2021-06-27 13:03:29+00:00   en  e385d4d639c6a423   \n",
       "38608  2021-06-27 08:37:21+00:00   en  257640324f249a73   \n",
       "38609  2021-06-26 08:03:22+00:00   en  000b71538f35fe46   \n",
       "38610  2021-06-25 21:02:17+00:00   en  08f94128283f7bc6   \n",
       "\n",
       "       public_metrics.retweet_count  public_metrics.reply_count  \\\n",
       "0                               0.0                         0.0   \n",
       "1                               0.0                         0.0   \n",
       "2                               0.0                         0.0   \n",
       "3                               0.0                         0.0   \n",
       "4                               0.0                         1.0   \n",
       "...                             ...                         ...   \n",
       "38606                           0.0                         1.0   \n",
       "38607                           0.0                         1.0   \n",
       "38608                           0.0                         1.0   \n",
       "38609                           0.0                         0.0   \n",
       "38610                           8.0                         6.0   \n",
       "\n",
       "       public_metrics.like_count  public_metrics.quote_count  \\\n",
       "0                            0.0                         0.0   \n",
       "1                            0.0                         0.0   \n",
       "2                            0.0                         0.0   \n",
       "3                            4.0                         0.0   \n",
       "4                            0.0                         0.0   \n",
       "...                          ...                         ...   \n",
       "38606                        2.0                         0.0   \n",
       "38607                        6.0                         0.0   \n",
       "38608                       17.0                         0.0   \n",
       "38609                        1.0                         0.0   \n",
       "38610                       38.0                         6.0   \n",
       "\n",
       "       public_metrics.impression_count  ...   Neutral  Positive     CLASS  \\\n",
       "0                                  0.0  ...  0.883804  0.048212   NEUTRAL   \n",
       "1                                  0.0  ...  0.583374  0.050595   NEUTRAL   \n",
       "2                                  0.0  ...  0.443080  0.253944   NEUTRAL   \n",
       "3                                  0.0  ...  0.076047  0.901351  POSITIVE   \n",
       "4                                  0.0  ...  0.275456  0.026258  NEGATIVE   \n",
       "...                                ...  ...       ...       ...       ...   \n",
       "38606                              0.0  ...  0.822089  0.142403   NEUTRAL   \n",
       "38607                              0.0  ...  0.093805  0.012996  NEGATIVE   \n",
       "38608                              0.0  ...  0.520120  0.323320   NEUTRAL   \n",
       "38609                              0.0  ...  0.063759  0.023034  NEGATIVE   \n",
       "38610                              0.0  ...  0.064148  0.012122  NEGATIVE   \n",
       "\n",
       "       num_CLASS           date_time  before_syrian_inflow  \\\n",
       "0              0 2016-04-20 22:55:08                 False   \n",
       "1              0 2016-04-20 21:27:37                 False   \n",
       "2              0 2016-04-20 21:18:58                 False   \n",
       "3              1 2016-04-20 18:25:11                 False   \n",
       "4             -1 2016-04-20 16:27:28                 False   \n",
       "...          ...                 ...                   ...   \n",
       "38606          0 2021-06-28 14:43:31                 False   \n",
       "38607         -1 2021-06-27 13:03:29                 False   \n",
       "38608          0 2021-06-27 08:37:21                 False   \n",
       "38609         -1 2021-06-26 08:03:22                 False   \n",
       "38610         -1 2021-06-25 21:02:17                 False   \n",
       "\n",
       "      before_ukrainian_inflow before_inflow during_inflow CLASS_bin  \n",
       "0                       False         False          True         0  \n",
       "1                       False         False          True         0  \n",
       "2                       False         False          True         0  \n",
       "3                       False         False          True         0  \n",
       "4                       False         False          True         1  \n",
       "...                       ...           ...           ...       ...  \n",
       "38606                    True          True         False         0  \n",
       "38607                    True          True         False         1  \n",
       "38608                    True          True         False         0  \n",
       "38609                    True          True         False         1  \n",
       "38610                    True          True         False         1  \n",
       "\n",
       "[38611 rows x 34 columns]"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "d2f2548f-5738-4a13-a3da-4cf086633844",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    20909\n",
       "1    17702\n",
       "Name: CLASS_bin, dtype: int64"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['CLASS_bin'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea2ddb39-ce25-4e1f-a398-462e78b36888",
   "metadata": {},
   "source": [
    "The value counts are as expected:\n",
    "* Negative: 17702\n",
    "* Not negative: 20909"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "83d79a08-a2af-4df7-8da9-fa64650489e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# checking for nulls (missings)\n",
    "\n",
    "df['CLASS_bin'].isnull().values.any()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42361f1d-c462-47bc-b070-9878f4eb53b2",
   "metadata": {},
   "source": [
    "### 5.1. Exploring CLASS_bin\n",
    "#### 5.1.1. Exploring CLASS_bin for entire dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "0bfbc75a-4086-4ebf-a514-8c95aba2b9a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.54153\n",
       "1    0.45847\n",
       "Name: CLASS_bin, dtype: float64"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the relative frequencies of CLASS\n",
    "\n",
    "df['CLASS_bin'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3e386de-fd63-4dca-ad0f-13bd9fa1d901",
   "metadata": {},
   "source": [
    "Approx. 54% of all tweets are not negative, while approx. 46% of all tweets are negative."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "ba685590-218c-4715-9c3d-ac4c3794be0b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: ylabel='CLASS_bin'>"
      ]
     },
     "execution_count": 145,
     "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": [
    "# pie chart\n",
    "\n",
    "# first color: redish hex code\n",
    "# second color: yellow-greenish hex code\n",
    "\n",
    "colors = '#E9F77D', '#EA4035'\n",
    "\n",
    "# adding percentages\n",
    "\n",
    "df.CLASS_bin.value_counts().plot(kind='pie', autopct='%1.0f%%', colors=colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e1bb548-06f0-4f36-b04d-43d71c86eebb",
   "metadata": {},
   "source": [
    "#### 5.1.2. Exploring CLASS_bin by inflow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "e7be7b27-f209-43e8-b5a9-7c5c1c715e8f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "inflow      CLASS_bin\n",
       "Syrians     0            0.633573\n",
       "            1            0.366427\n",
       "Ukrainians  1            0.620566\n",
       "            0            0.379434\n",
       "Name: CLASS_bin, dtype: float64"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding the proportion of classes by inflow\n",
    "\n",
    "byinflow = df.groupby('inflow').CLASS_bin.value_counts(normalize=True)\n",
    "byinflow"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1271c815-5d97-4b88-b3c3-77190addc266",
   "metadata": {},
   "source": [
    "#### Remarks\n",
    "The proportion of negative and not negative tweets surrounding both inflows resemble one another."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8665577b-2831-491b-b651-6aecb709f2ea",
   "metadata": {},
   "source": [
    "### 5.2. Exploring by date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "6b387536-fdd3-44bb-890b-0305e3174375",
   "metadata": {},
   "outputs": [],
   "source": [
    "# re-establishing syrian & ukrainian dataset (to use binary CLASS)\n",
    "\n",
    "df_syr = df[df.inflow == 'Syrians'].copy()\n",
    "\n",
    "df_uk = df[df.inflow == 'Ukrainians'].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "64ad2234-b8be-4a2f-857c-b580acd5a892",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "before_inflow  CLASS_bin\n",
       "False          0            0.634907\n",
       "               1            0.365093\n",
       "True           0            0.627245\n",
       "               1            0.372755\n",
       "Name: CLASS_bin, dtype: float64"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding the proportion of classes by before or during syrian inflow\n",
    "\n",
    "bybord = df_syr.groupby('before_inflow').CLASS_bin.value_counts(normalize=True)\n",
    "bybord"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "301878b0-7265-4219-9c6e-ff91b89c2fc4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "before_inflow  CLASS_bin\n",
       "False          1            0.618371\n",
       "               0            0.381629\n",
       "True           1            0.635405\n",
       "               0            0.364595\n",
       "Name: CLASS_bin, dtype: float64"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# finding the proportion of classes by before or during ukrainian inflow\n",
    "\n",
    "bybord = df_uk.groupby('before_inflow').CLASS_bin.value_counts(normalize=True)\n",
    "bybord"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc7f5835-9a17-49d4-9db4-1ef5322ff223",
   "metadata": {},
   "source": [
    "### 5.3 Saving the data\n",
    "Finally, the dataframe containing negative or not negative tweets is saved."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "1ffadc10-ec7c-4b01-9cc4-6b805ecf0c6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(CASS_thesis / '05_Explored-and-vizualized_limited_merged_binary-CLASS.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8f4a335-e9b2-4f89-aca2-cc73977e9ecc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}