youtube_presidentielles_code.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6922d4b5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/heloise/.pyenv/versions/3.10.0/envs/youtube_presidentielles/lib/python3.10/site-packages/secretstorage/util.py:46: UserWarning: Passing unwrap= to .send_and_get_reply() is deprecated and will break in a future version of Jeepney.\n",
      "  return self._connection.send_and_get_reply(msg, unwrap=True)\n"
     ]
    }
   ],
   "source": [
    "import minet\n",
    "import pandas as pd\n",
    "import requests\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import lines\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b965e68",
   "metadata": {},
   "source": [
    "# Installation"
   ]
  },
  {
   "cell_type": "raw",
   "id": "eb153660",
   "metadata": {},
   "source": [
    "The scripts were run on Python 3.10.0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2c57cf65",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install -r requirements.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1c2593a",
   "metadata": {},
   "source": [
    "# Data collection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1d94a27d",
   "metadata": {},
   "outputs": [],
   "source": [
    "!minet youtube search macron --order date > macron_videos_raw.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "79157997",
   "metadata": {},
   "outputs": [],
   "source": [
    "!minet youtube search \"le pen\" --order date > lepen_videos_raw.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0df83621",
   "metadata": {},
   "source": [
    "# Data preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "aef19395",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1 = pd.read_csv('./macron_videos_raw.csv')\n",
    "df_1 = df_1.drop_duplicates(subset=['video_id'])\n",
    "df_1 = df_1.head(500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a0cc8704",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_information_panel_data(df):\n",
    "    df['video_url'] = \"https://www.youtube.com/watch?v=\" + df['video_id']\n",
    "    df['election_panel'] = df['video_url'].apply(lambda x: 'fr2022_president_erip' in requests.get(x).text)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bdafcd2b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1 = get_information_panel_data(df_1)\n",
    "df_1.to_csv('./macron_videos.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "58733961",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_2 = pd.read_csv('./lepen_videos_raw.csv')\n",
    "df_2 = df_2.drop_duplicates(subset=['video_id'])\n",
    "df_2 = df_2.head(500)\n",
    "df_2 = get_information_panel_data(df_2)\n",
    "df_2.to_csv('./lepen_videos.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "793886f7",
   "metadata": {},
   "source": [
    "# Data vizualisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "14a4db78",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1 = pd.read_csv('./macron_videos.csv')\n",
    "df_1['published_at'] = pd.to_datetime(df_1['published_at'])\n",
    "\n",
    "df_2 = pd.read_csv('./lepen_videos.csv')\n",
    "df_2['published_at'] = pd.to_datetime(df_2['published_at'])\n",
    "\n",
    "df_t = pd.concat([df_1, df_2])\n",
    "df_t = df_t.drop_duplicates(subset=['video_id'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fd34eb3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 612x108 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8.5, 1.5))\n",
    "ax = plt.subplot(111)\n",
    "\n",
    "for date in df_t[~df_t['election_panel']]['published_at'].values:\n",
    "    plt.plot([date, date], [0, 1], color='C2', alpha=0.5)\n",
    "for date in df_t[df_t['election_panel']]['published_at'].values:\n",
    "    plt.plot([date, date], [0, 1], color='C1', alpha=0.5)\n",
    "    \n",
    "plt.xlim(np.datetime64('2021-12-16'), np.datetime64('2022-04-15'))\n",
    "plt.xticks(\n",
    "    ticks=[np.datetime64(x) for x in ['2022-01-01', '2022-02-01', '2022-03-01', '2022-04-01']], \n",
    "    labels=['1er janvier 2022', '1er février 2022', '1er mars 2022', '1er avril 2022']\n",
    ")\n",
    "plt.yticks([])\n",
    "vertical_line_1 = lines.Line2D([], [], color='C2', marker='|', linestyle='None',\n",
    "                          markersize=15, markeredgewidth=1.5, label=\"Vidéo sans panneau d'information\")\n",
    "vertical_line_2 = lines.Line2D([], [], color='C1', marker='|', linestyle='None',\n",
    "                          markersize=15, markeredgewidth=1.5, label=\"Vidéo avec panneau d'information\")\n",
    "plt.legend(handles=[vertical_line_1, vertical_line_2], ncol=2, loc='upper center', frameon=False)\n",
    "plt.ylim(0, 1.8)\n",
    "ax.set_frame_on(False)\n",
    "plt.tight_layout()\n",
    "plt.savefig('./figure_2.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "35b32993",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "177"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_t[df_t['published_at'] < '2022-03-10 00:00:00'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a4115963",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6, 3), subplot_kw=dict(aspect=\"equal\"))\n",
    "\n",
    "def func(pct, allvals):\n",
    "    absolute = int(np.round(pct/100.*np.sum(allvals)))\n",
    "    return \"{:d}\\nvidéos\\n({:.1f}%)\".format(absolute, pct)\n",
    "\n",
    "df_1b = df_1[df_1['published_at'] > '2022-03-10 00:00:00']\n",
    "plt.subplot(121)\n",
    "data = [len(df_1b[df_1b['election_panel']]), len(df_1b[~df_1b['election_panel']])]\n",
    "labels = [\"avec panneau\\nd'information\", \"sans panneau\\nd'information\"]\n",
    "wedges, texts, autotexts = plt.pie(data, colors=['C1', 'C2'],\n",
    "                                  autopct=lambda pct: func(pct, data),\n",
    "                                  textprops=dict(color=\"w\"))\n",
    "plt.title(\"Vidéos sur E. Macron\\ndepuis le 10 mars 2022\")\n",
    "\n",
    "df_2b = df_2[df_2['published_at'] > '2022-03-10 00:00:00']\n",
    "plt.subplot(122)\n",
    "data = [len(df_2b[df_2b['election_panel']]), len(df_2b[~df_2b['election_panel']])]\n",
    "wedges, texts, autotexts = plt.pie(data, colors=['C1', 'C2'],\n",
    "                                  autopct=lambda pct: func(pct, data),\n",
    "                                  textprops=dict(color=\"w\"))\n",
    "plt.legend(wedges, labels, framealpha=0,\n",
    "          loc=\"center left\", bbox_to_anchor=(1, 0, 0.5, 1))\n",
    "plt.title(\"Vidéos sur M. Le Pen\\ndepuis le 10 mars 2022\")\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig('./figure_3.png')"
   ]
  }
 ],
 "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": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}