{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sujet 5 - Analyse des dialogues dans l'Avare de Molière"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Récupérer les données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import re"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "L’Observatoire de la vie littéraire (OBVIL) promeut une approche de l'analyse des textes littéraires fondée sur le numérique. Dans le cadre du Projet Molière, des pièces de cet auteur ont été numérisées et sont accessibles librement dans différents formats utilisables par un programme informatique. Nous allons utiliser ici les textes sous format markdown accessibles [ici](http://dramacode.github.io/markdown/moliere_avare.txt)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_url = 'http://dramacode.github.io/markdown/moliere_avare.txt'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pour plus de reproductibilité, nous allons télécharger les données dans ce répertoire GitLab d'abord puis nous allons lire ce fichier plutôt que l'url directement."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_file = \"moliere_avare.txt\"\n",
    "\n",
    "import os\n",
    "import urllib.request\n",
    "if not os.path.exists(data_file):\n",
    "    urllib.request.urlretrieve(data_url, data_file)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous pouvons regarder les premières lignes de ce fichier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext rpy2.ipython"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---\n",
      "identifier: moliere_avare  \n",
      "creator: Molière.  \n",
      "date: 1668  \n",
      "title: L'Avare. Comédie  \n",
      "---\n",
      "\n",
      "\n",
      "L'AVARE,\n",
      "\n",
      "COMÉDIE.\n",
      "\n",
      "Par J.B.P. MOLIÈRE.\n",
      "\n",
      "À PARIS, Chez JEAN RIBOU, au Palais, vis à vis la Porte de l'Église de la Sainte Chapelle, à l'Image Saint-Louis. M. DC. LXIX. *AVEC PRIVILÈGE DU ROI*\n",
      "\n",
      "\n",
      "\n",
      "# ACTEURS.\n",
      " – Harpagon, Père de Cléante et d'Élise, et Amoureux de Mariane.\n",
      " – Cléante, Fils d'Harpagon, Amant de Mariane.\n",
      " – Élise, Fille d'Harpagon, Amante de Valère.\n",
      " – Valère, Fils d'Anselme, et Amant d'Élise.\n",
      " – Mariane, Amante de Cléante, et aimée d'Harpagon.\n",
      " – Anselme, Père de Valère et de Mariane.\n",
      " – Frosine, Femme d'Intrigue.\n",
      " – Maitre Simon, Courtier.\n",
      " – Maitre Jacques, Cuisinier et Cocher d'Harpagon.\n",
      " – La Flèche, Valet de Cléante.\n",
      " – Dame Claude, Servante d'Harpagon.\n",
      " – Brindavoine, laquais d'Harpagon.\n",
      " – La Merluche, laquais d'Harpagon.\n",
      " – Le commissaire, et son clerc.\n",
      "La Scène est à Paris.\n",
      "\n",
      "\n",
      "\n",
      "# L'Avare, *Comédie.*.\n",
      "\n",
      "\n",
      "## Acte Premier.\n",
      "\n",
      "\n",
      "### Scène Première.\n",
      "Valère, Élise\n",
      "\n",
      "\n",
      "    VALÈRE.\n",
      "Hé quoi, charmante Élise, vous devenez mélancolique, après les obligeantes assurances que vous avez eu la bonté de me donner de votre foi ?Je vous vois soupirer, hélas, au milieu de ma joie !Est-ce du regret, dites-moi, de m'avoir fait heureux ? et vous repentez-vous de cet engagement où mes feux ont pu vous contraindre ?\n",
      "\n",
      "    ÉLISE.\n",
      "Non, Valère, je ne puis pas me repentir de tout ce que je fais pour vous. Je m'y sens entraîner par une trop douce puissance, et je n'ai pas même la force de souhaiter que les choses ne fussent pas. Mais, à vous dire vrai, le succès me donne de l'inquiétude ; et je crains fort de vous aimer un peu plus que je ne devrais.\n",
      "\n",
      "    VALÈRE.\n",
      "Hé que pouvez-vous craindre, Élise, dans les bontés que vous avez pour moi ?\n"
     ]
    }
   ],
   "source": [
    "%%sh\n",
    "head -n 55 moliere_avare.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comme nous pouvons le voir, les actes sont marqués par un double-dièse en début de ligne et les scènes sont marquées par un triple-dièse en début de ligne. Les dialogues de personnages ont l'air d'être sur 2 lignes : une première avec le nom du personnage en majuscule et une deuxième avec les répliques du personnage.\n",
    "\n",
    "Nous allons tenter de réarranger les données sous forme de tableau, comme ceci :\n",
    "\n",
    "| Personnage | Acte | Scene | Nombre de Mots | Nombre de Repliques |\n",
    "|:-------|:------|:-------|:------------|:----------------|\n",
    "| nom du personnage | acte dans lequel il apparait | scène dans laquelle il figure | le nombre de mots qu'il parle | le nombre de repliques |\n",
    "\n",
    "Nous allons créer un fonction qui va remplacer les caractères à accent en caractère \"normaux\":"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import unicodedata\n",
    "\n",
    "def remove_accents(input_str):\n",
    "    nfkd_form = unicodedata.normalize('NFKD', input_str)\n",
    "    return u\"\".join([c for c in nfkd_form if not unicodedata.combining(c)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comme nous pouvons le voir, il y a quelques discordances entre la liste des personnages énumérés en-dessous du numéro de scène et les répliques dans le dialogue."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VALERE is not listed under scene 19 acte 3 but is speaking\n",
      "No characters listed for scene 26 acte 4\n",
      "HARPAGON is not listed under scene 26 acte 4 but is speaking\n"
     ]
    },
    {
     "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>acte</th>\n",
       "      <th>nombre_de_mots</th>\n",
       "      <th>nombre_de_repliques</th>\n",
       "      <th>personnage</th>\n",
       "      <th>scene</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>596</td>\n",
       "      <td>8</td>\n",
       "      <td>VALERE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>473</td>\n",
       "      <td>8</td>\n",
       "      <td>ELISE</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>725</td>\n",
       "      <td>10</td>\n",
       "      <td>CLEANTE</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>150</td>\n",
       "      <td>9</td>\n",
       "      <td>ELISE</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>396</td>\n",
       "      <td>34</td>\n",
       "      <td>HARPAGON</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   acte  nombre_de_mots  nombre_de_repliques personnage  scene\n",
       "0     1             596                    8     VALERE      1\n",
       "1     1             473                    8      ELISE      1\n",
       "2     1             725                   10    CLEANTE      2\n",
       "3     1             150                    9      ELISE      2\n",
       "4     1             396                   34   HARPAGON      3"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acte = 0\n",
    "scene = 0\n",
    "infos_scene = {}\n",
    "\n",
    "data = []\n",
    "with open(data_file) as f:\n",
    "    lines = f.readlines()\n",
    "nbline = 0\n",
    "while nbline < len(lines):\n",
    "    line = lines[nbline]\n",
    "    if line.startswith(\"## Acte\"):\n",
    "        acte += 1\n",
    "    elif line.startswith(\"### Sc\"):\n",
    "        if infos_scene:\n",
    "            data += infos_scene.values()\n",
    "        scene += 1\n",
    "        nbline += 1\n",
    "        line = lines[nbline]\n",
    "        if line.strip():\n",
    "            infos_scene = {l.strip().upper():{'acte':acte,'scene':scene,'personnage':l.strip().upper(),'nombre_de_mots':0,'nombre_de_repliques':0} for l in remove_accents(line.strip()).split(\",\")}\n",
    "        else:\n",
    "            infos_scene = {}\n",
    "            print(\"No characters listed for scene\",scene,\"acte\",acte)\n",
    "    elif re.search(r\"^    [A-ZÈÉ ]+.$\",line):\n",
    "        assert acte and scene\n",
    "        personnage = remove_accents(re.search(r\"^    ([A-ZÈÉ ]+).$\",line).groups()[0])\n",
    "        nbline += 1\n",
    "        line = lines[nbline]\n",
    "        assert line.strip() # check line is not empty\n",
    "        nombre_de_mots = len(line.split()) # on va supposer que la ponctuation est négligeable dans le compte\n",
    "        if personnage not in infos_scene:\n",
    "            print(personnage,\"is not listed under scene\",scene,\"acte\",acte,\"but is speaking\")\n",
    "            infos_scene[personnage] = {'acte':acte,'scene':scene,'personnage':personnage,'nombre_de_mots':0,'nombre_de_repliques':0}\n",
    "        infos_scene[personnage]['nombre_de_repliques'] += 1\n",
    "        infos_scene[personnage]['nombre_de_mots'] += nombre_de_mots\n",
    "    nbline += 1\n",
    "df = pd.DataFrame(data)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Voyons voir s'il y a bien le bon nombre d'Actes et de Scènes:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5\n",
      "32\n"
     ]
    }
   ],
   "source": [
    "%%sh\n",
    "grep -c \"## Acte\" moliere_avare.txt\n",
    "grep -c \"### Scène\" moliere_avare.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5,)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.acte.value_counts().shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(31,)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.scene.value_counts().shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15,)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.personnage.value_counts().shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On voit donc qu'on retrouve le même nombre d'actes et de scènes qu'avec l'analyse de texte via `grep`. On retrouve aussi les 15 personnages listées en début de document."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " – Harpagon, Père de Cléante et d'Élise, et Amoureux de Mariane.\n",
      " – Cléante, Fils d'Harpagon, Amant de Mariane.\n",
      " – Élise, Fille d'Harpagon, Amante de Valère.\n",
      " – Valère, Fils d'Anselme, et Amant d'Élise.\n",
      " – Mariane, Amante de Cléante, et aimée d'Harpagon.\n",
      " – Anselme, Père de Valère et de Mariane.\n",
      " – Frosine, Femme d'Intrigue.\n",
      " – Maitre Simon, Courtier.\n",
      " – Maitre Jacques, Cuisinier et Cocher d'Harpagon.\n",
      " – La Flèche, Valet de Cléante.\n",
      " – Dame Claude, Servante d'Harpagon.\n",
      " – Brindavoine, laquais d'Harpagon.\n",
      " – La Merluche, laquais d'Harpagon.\n",
      " – Le commissaire, et son clerc.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\".join(lines[19:33]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyser les données"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Classons les personnages selon la quantité de parole grâce à une analyse syntaxique du texte (scènes / répliques / mots). En particulier, quel est celui qui parle le plus ? Quel est celui qui ne parle pas du tout ?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "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>personnage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>HARPAGON</th>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FROSINE</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CLEANTE</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ELISE</th>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIANE</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VALERE</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAITRE JACQUES</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LA FLECHE</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LE COMMISSAIRE</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SON CLERC</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BRINDAVOINE</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LA MERLUCHE</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DAME CLAUDE</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANSELME</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAITRE SIMON</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                personnage\n",
       "HARPAGON                22\n",
       "FROSINE                 14\n",
       "CLEANTE                 14\n",
       "ELISE                   13\n",
       "MARIANE                 11\n",
       "VALERE                   8\n",
       "MAITRE JACQUES           8\n",
       "LA FLECHE                5\n",
       "LE COMMISSAIRE           5\n",
       "SON CLERC                5\n",
       "BRINDAVOINE              2\n",
       "LA MERLUCHE              2\n",
       "DAME CLAUDE              1\n",
       "ANSELME                  1\n",
       "MAITRE SIMON             1"
      ]
     },
     "execution_count": 179,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(df.personnage.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>nombre_de_mots</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>personnage</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DAME CLAUDE</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAITRE JACQUES</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAITRE SIMON</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SON CLERC</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BRINDAVOINE</th>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LA MERLUCHE</th>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LE COMMISSAIRE</th>\n",
       "      <td>258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANSELME</th>\n",
       "      <td>383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIANE</th>\n",
       "      <td>819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ELISE</th>\n",
       "      <td>893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LA FLECHE</th>\n",
       "      <td>1419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FROSINE</th>\n",
       "      <td>2033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VALERE</th>\n",
       "      <td>2532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CLEANTE</th>\n",
       "      <td>3046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HARPAGON</th>\n",
       "      <td>5092</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                nombre_de_mots\n",
       "personnage                    \n",
       "DAME CLAUDE                  0\n",
       "MAITRE JACQUES               0\n",
       "MAITRE SIMON                 0\n",
       "SON CLERC                    0\n",
       "BRINDAVOINE                 38\n",
       "LA MERLUCHE                 49\n",
       "LE COMMISSAIRE             258\n",
       "ANSELME                    383\n",
       "MARIANE                    819\n",
       "ELISE                      893\n",
       "LA FLECHE                 1419\n",
       "FROSINE                   2033\n",
       "VALERE                    2532\n",
       "CLEANTE                   3046\n",
       "HARPAGON                  5092"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['nombre_de_mots','personnage']].groupby('personnage').sum().sort_values(by=['nombre_de_mots'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>nombre_de_repliques</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>personnage</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DAME CLAUDE</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAITRE JACQUES</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAITRE SIMON</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SON CLERC</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BRINDAVOINE</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LA MERLUCHE</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANSELME</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LE COMMISSAIRE</th>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIANE</th>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ELISE</th>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FROSINE</th>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LA FLECHE</th>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VALERE</th>\n",
       "      <td>99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CLEANTE</th>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HARPAGON</th>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                nombre_de_repliques\n",
       "personnage                         \n",
       "DAME CLAUDE                       0\n",
       "MAITRE JACQUES                    0\n",
       "MAITRE SIMON                      0\n",
       "SON CLERC                         0\n",
       "BRINDAVOINE                       3\n",
       "LA MERLUCHE                       5\n",
       "ANSELME                          14\n",
       "LE COMMISSAIRE                   15\n",
       "MARIANE                          26\n",
       "ELISE                            50\n",
       "FROSINE                          59\n",
       "LA FLECHE                        64\n",
       "VALERE                           99\n",
       "CLEANTE                         156\n",
       "HARPAGON                        334"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['nombre_de_repliques','personnage']].groupby('personnage').sum().sort_values(by=['nombre_de_repliques'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On voit dans ces analyses qu'Harpagon participe au plus grand nombre de scènes (22 sur 31). En terme de nombre de mots parlés, Harpagon est aussi celui qui parle le plus avec Dame Claude, Maitre Jacques, Maitre Simon et le clerc qui ne parlent pas du tout. C'est aussi Harpagon qui a le plus grand nombre de répliques."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous allons maintenant représenter la distribution des nombres de mots parlés par personne dans chaque scène sous forme de graphique. Pour cela, nous allons tout d'abord mettre en forme les données en extrayant les colonnes qui nous intéressent et en les formatant de \"long\" à \"wide\" (2D) compatible à pd.plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "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>scene</th>\n",
       "      <th>personnage</th>\n",
       "      <th>nombre_de_mots</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>5</td>\n",
       "      <td>VALERE</td>\n",
       "      <td>621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>CLEANTE</td>\n",
       "      <td>725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>6</td>\n",
       "      <td>LA FLECHE</td>\n",
       "      <td>853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>4</td>\n",
       "      <td>HARPAGON</td>\n",
       "      <td>1044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>10</td>\n",
       "      <td>FROSINE</td>\n",
       "      <td>1234</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    scene personnage  nombre_de_mots\n",
       "11      5     VALERE             621\n",
       "2       2    CLEANTE             725\n",
       "13      6  LA FLECHE             853\n",
       "8       4   HARPAGON            1044\n",
       "22     10    FROSINE            1234"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_wc = df[['scene','personnage','nombre_de_mots']].groupby(['scene','personnage']).sum().reset_index()\n",
    "df_wc.sort_values(by='nombre_de_mots').tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "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>personnage</th>\n",
       "      <th>scene</th>\n",
       "      <th>ANSELME</th>\n",
       "      <th>BRINDAVOINE</th>\n",
       "      <th>CLEANTE</th>\n",
       "      <th>DAME CLAUDE</th>\n",
       "      <th>ELISE</th>\n",
       "      <th>FROSINE</th>\n",
       "      <th>HARPAGON</th>\n",
       "      <th>LA FLECHE</th>\n",
       "      <th>LA MERLUCHE</th>\n",
       "      <th>LE COMMISSAIRE</th>\n",
       "      <th>MAITRE JACQUES</th>\n",
       "      <th>MAITRE SIMON</th>\n",
       "      <th>MARIANE</th>\n",
       "      <th>SON CLERC</th>\n",
       "      <th>VALERE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>473.0</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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>596.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>725.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>150.0</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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</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>0.0</td>\n",
       "      <td>396.0</td>\n",
       "      <td>242.0</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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>211.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>147.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1044.0</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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>238.0</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>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>621.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "personnage  scene  ANSELME  BRINDAVOINE  CLEANTE  DAME CLAUDE  ELISE  FROSINE  \\\n",
       "0               1      0.0          0.0      0.0          0.0  473.0      0.0   \n",
       "1               2      0.0          0.0    725.0          0.0  150.0      0.0   \n",
       "2               3      0.0          0.0      0.0          0.0    0.0      0.0   \n",
       "3               4      0.0          0.0    211.0          0.0  147.0      0.0   \n",
       "4               5      0.0          0.0      0.0          0.0   36.0      0.0   \n",
       "\n",
       "personnage  HARPAGON  LA FLECHE  LA MERLUCHE  LE COMMISSAIRE  MAITRE JACQUES  \\\n",
       "0                0.0        0.0          0.0             0.0             0.0   \n",
       "1                0.0        0.0          0.0             0.0             0.0   \n",
       "2              396.0      242.0          0.0             0.0             0.0   \n",
       "3             1044.0        0.0          0.0             0.0             0.0   \n",
       "4              238.0        0.0          0.0             0.0             0.0   \n",
       "\n",
       "personnage  MAITRE SIMON  MARIANE  SON CLERC  VALERE  \n",
       "0                    0.0      0.0        0.0   596.0  \n",
       "1                    0.0      0.0        0.0     0.0  \n",
       "2                    0.0      0.0        0.0     0.0  \n",
       "3                    0.0      0.0        0.0     0.0  \n",
       "4                    0.0      0.0        0.0   621.0  "
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2d_wc = df_wc.pivot('scene', 'personnage', 'nombre_de_mots').fillna(0).reset_index()  #Reshape from long to wide\n",
    "df2d_wc.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df2d_wc.plot(x = 'scene',\n",
    "    kind = 'barh',\n",
    "    stacked = True,\n",
    "    title = 'Number of words per scene and per character',\n",
    "    mark_right = True)\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now plot the interactions between characters as a network where nodes are characters and edges are plotted if these characters are together in a scene."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [],
   "source": [
    "edges = {}\n",
    "for scene in df.scene.unique():\n",
    "    personnages = sorted(list(df[df.scene==scene].personnage))\n",
    "    for p1 in range(0,len(personnages)-1):\n",
    "        for p2 in range(p1+1,len(personnages)):\n",
    "            if personnages[p1] not in edges:\n",
    "                edges[personnages[p1]] = {}\n",
    "            if personnages[p2] not in edges[personnages[p1]]:\n",
    "                edges[personnages[p1]][personnages[p2]] = 0\n",
    "            edges[personnages[p1]][personnages[p2]] += 1\n",
    "edges_tuples = []\n",
    "for p1,d in edges.items():\n",
    "    for p2,count in d.items():\n",
    "        edges_tuples.append([p1,p2,count])\n",
    "df_edges = pd.DataFrame(edges_tuples, columns=['source','target','weight'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "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>source</th>\n",
       "      <th>target</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>CLEANTE</td>\n",
       "      <td>ELISE</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>FROSINE</td>\n",
       "      <td>HARPAGON</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ELISE</td>\n",
       "      <td>HARPAGON</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CLEANTE</td>\n",
       "      <td>HARPAGON</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>FROSINE</td>\n",
       "      <td>MARIANE</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     source    target  weight\n",
       "8   CLEANTE     ELISE       8\n",
       "27  FROSINE  HARPAGON      10\n",
       "1     ELISE  HARPAGON      10\n",
       "9   CLEANTE  HARPAGON      10\n",
       "30  FROSINE   MARIANE      11"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_edges.sort_values(by='weight').tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [],
   "source": [
    "import networkx as nx\n",
    "g = nx.Graph()\n",
    "g = nx.from_pandas_edgelist(df_edges, edge_attr=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nx.draw(g, with_labels = True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}