{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c00",
   "metadata": {},
   "source": [
    "# 09 — Calibration : estimer γα/λ sur des données publiques\n",
    "\n",
    "C'est le chiffre qui manquait.\n",
    "\n",
    "Le [critère d'instabilité](../theorie/resonance.md) $\\gamma\\alpha > \\lambda$ est la\n",
    "recommandation la plus opérationnelle du [mémorandum](../memorandum.md) : au-delà de ce\n",
    "seuil, l'amplification d'un contenu excède son taux d'oubli naturel et le système accumule\n",
    "l'énergie qu'on lui injecte. Mais jusqu'ici, **personne ne savait où se situent les\n",
    "systèmes réels.** L'inégalité était formelle, donc inopposable.\n",
    "\n",
    "Ce notebook l'estime sur des séries d'attention publiques.\n",
    "\n",
    "**Résultat, annoncé d'emblée :** sur 19 épisodes d'attention mesurés, le rapport\n",
    "$\\gamma\\alpha/\\lambda$ vaut entre **1,5 et 12**, de médiane **2,5 à 4,2** selon\n",
    "l'estimateur. Il est supérieur à 1 dans **tous** les épisodes. Deux conséquences, qui vont\n",
    "en sens contraire :\n",
    "\n",
    "* l'ordre de grandeur existe désormais — l'amplification est deux à quatre fois plus rapide\n",
    "  que l'oubli ;\n",
    "* **le critère de signe est vide.** Il est satisfait par construction pour tout épisode\n",
    "  observable, et la recommandation du mémorandum doit être reformulée en **plafond sur le\n",
    "  rapport**, non en vérification de signe.\n",
    "\n",
    "Et une prédiction du modèle n'est **pas** vérifiée : les contenus à forte charge\n",
    "émotionnelle ne montrent pas de rapport plus élevé que les annonces scientifiques."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c01",
   "metadata": {},
   "source": [
    "## 1. Ce qui est estimé, et comment\n",
    "\n",
    "L'équation complète décrit un oscillateur, et son contenu oscillatoire n'a de sens que dans\n",
    "le régime de cycle limite. Pour un **pic d'attention isolé**, la dynamique se réduit à sa\n",
    "forme du premier ordre :\n",
    "\n",
    "$$\\frac{dV}{dt} = \\gamma\\alpha\\,\\sigma(V)\\,V - \\lambda V$$\n",
    "\n",
    "Elle se sépare alors en deux régimes directement mesurables :\n",
    "\n",
    "| Phase | Condition | Comportement | Pente mesurée |\n",
    "|---|---|---|---|\n",
    "| **montée** | $V \\ll V_{\\text{sat}}$, donc $\\sigma \\approx 1$ | $V \\propto e^{(\\gamma\\alpha-\\lambda)t}$ | $r_{\\text{up}} = \\gamma\\alpha - \\lambda$ |\n",
    "| **décroissance** | saturation atteinte, déclencheur passé | $V \\propto e^{-\\lambda t}$ | $r_{\\text{down}} = \\lambda$ |\n",
    "\n",
    "D'où l'identification, par deux régressions log-linéaires sur un même épisode :\n",
    "\n",
    "$$\\lambda = r_{\\text{down}}, \\qquad\n",
    "  \\gamma\\alpha = r_{\\text{up}} + r_{\\text{down}}, \\qquad\n",
    "  \\boxed{\\frac{\\gamma\\alpha}{\\lambda} = 1 + \\frac{r_{\\text{up}}}{r_{\\text{down}}}}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c02",
   "metadata": {},
   "source": [
    "## 2. La source, et ce qu'elle n'est pas\n",
    "\n",
    "Les séries proviennent de l'API de consultations de Wikimedia — la seule source qui réunisse\n",
    "accès libre, granularité quotidienne, profondeur depuis 2015 et stabilité. Les archives\n",
    "Reddit ont fermé, l'API de X est devenue payante, Google Trends masque l'échelle absolue.\n",
    "\n",
    "Trois réserves doivent être posées **avant** de regarder les chiffres, parce qu'elles\n",
    "bornent ce qu'on peut en conclure :\n",
    "\n",
    "1. **Wikipédia n'a pas d'algorithme de recommandation.** Le $\\gamma$ estimé est le gain\n",
    "   composite de l'écosystème informationnel — recherche, partage, reprise médiatique — et\n",
    "   non la fonction de classement d'une plateforme. C'est une borne écosystémique, pas un\n",
    "   audit de plateforme.\n",
    "2. **Une consultation n'est pas une exposition.** Le modèle décrit ce qui est *servi* ; ces\n",
    "   séries mesurent ce qui est *consulté*, un observable en aval.\n",
    "3. **Le filtre d'agent est décisif.** Sans exclusion des robots, l'article OSIRIS-REx\n",
    "   présente un pic à 17 millions de consultations en un jour, sans rapport avec un épisode\n",
    "   d'attention réel. Toutes les séries employées ici sont filtrées sur l'agent `user`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c03",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:37:33.387721Z",
     "iopub.status.busy": "2026-08-23T13:37:33.387322Z",
     "iopub.status.idle": "2026-08-23T13:37:33.940213Z",
     "shell.execute_reply": "2026-08-23T13:37:33.939782Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "agent filtré        : user\n",
      "sujets pré-enregistrés : 24 (12 accusation, 12 découverte)\n",
      "fenêtre             : 2015-07-01 → 2026-08-01\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n",
    "\n",
    "from ide.calibration import EpisodeCriteria, scan_series\n",
    "from ide.corpus import CORPUS, CORPUS_END, CORPUS_START, by_category\n",
    "from ide.pageviews import DEFAULT_AGENT, load_or_fetch\n",
    "from ide.plotting import PALETTE, save_figure, use_project_style\n",
    "\n",
    "use_project_style()\n",
    "\n",
    "# Le cache est versionné : l'analyse est reproductible hors ligne, et un résultat publié\n",
    "# ne dépend pas de la disponibilité future d'un service tiers.\n",
    "SERIES = {\n",
    "    entry.label: (entry, load_or_fetch(entry.project, entry.article, CORPUS_START, CORPUS_END))\n",
    "    for entry in CORPUS\n",
    "}\n",
    "\n",
    "print(f\"agent filtré        : {DEFAULT_AGENT}\")\n",
    "print(f\"sujets pré-enregistrés : {len(CORPUS)} \"\n",
    "      f\"({len(by_category('accusation'))} accusation, {len(by_category('discovery'))} découverte)\")\n",
    "print(f\"fenêtre             : {CORPUS_START} → {CORPUS_END}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c04",
   "metadata": {},
   "source": [
    "## 3. Le corpus est pré-enregistré\n",
    "\n",
    "La comparaison entre classes n'a de valeur que si la liste des sujets est arrêtée **avant**\n",
    "de regarder les résultats. Elle est donc figée dans `ide.corpus`, et la règle est explicite :\n",
    "**aucun article n'est retiré au vu de son résultat.** Les sujets sans épisode exploitable\n",
    "sont rapportés comme tels.\n",
    "\n",
    "Les deux classes portent sur la **nature de l'émotion mobilisée**, non sur l'importance du\n",
    "sujet :\n",
    "\n",
    "* **accusation** — attention mobilisée par une accusation, une menace, un scandale : colère,\n",
    "  indignation, peur. Registre à $\\alpha$ élevé selon le modèle.\n",
    "* **découverte** — attention mobilisée par une découverte ou une réussite **non\n",
    "  programmée** : curiosité, admiration. Registre à $\\alpha$ faible.\n",
    "\n",
    "La restriction aux annonces non programmées vise un confondant précis : un événement soudain\n",
    "monte plus raide qu'un événement anticipé, quelle que soit sa charge émotionnelle. Si la\n",
    "classe « découverte » ne contenait que des remises de prix et des lancements annoncés de\n",
    "longue date, la prédiction serait vérifiée pour une raison de calendrier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c05",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:37:33.941099Z",
     "iopub.status.busy": "2026-08-23T13:37:33.940989Z",
     "iopub.status.idle": "2026-08-23T13:37:34.479479Z",
     "shell.execute_reply": "2026-08-23T13:37:34.478923Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "classe      sujet                        ép.  cand.  motif de rejet dominant\n",
      "----------------------------------------------------------------------------\n",
      "accusation  QAnon                          1     41  —\n",
      "accusation  Pizzagate                      1     25  —\n",
      "accusation  Cambridge Analytica            1     29  —\n",
      "accusation  Panama Papers                  1     25  —\n",
      "accusation  Paradise Papers                0     19  fenêtre\n",
      "accusation  Chemtrails                     1      8  —\n",
      "accusation  Grand remplacement             0     27  forme\n",
      "accusation  Désinformation Covid-19        0      6  forme\n",
      "accusation  Hésitation vaccinale           0      7  forme\n",
      "accusation  Pegasus (logiciel espion)      0     40  fenêtre\n",
      "accusation  Affaire Benalla                1     45  —\n",
      "accusation  Gilets jaunes                  1     16  —\n",
      "discovery   Ondes gravitationnelles        1     17  —\n",
      "discovery   LIGO                           2     14  —\n",
      "discovery   Event Horizon Telescope        0     36  forme\n",
      "discovery   Télescope James-Webb           2     16  —\n",
      "discovery   CRISPR                         0      5  fenêtre\n",
      "discovery   Boson de Higgs                 1     10  —\n",
      "discovery   Perseverance                   2     18  —\n",
      "discovery   Sursauts radio rapides         3     62  —\n",
      "discovery   AlphaFold                      0     20  trafic\n",
      "discovery   OSIRIS-REx                     0     28  forme\n",
      "discovery   Trou noir                      1      8  —\n",
      "discovery   Télescope James-Webb (fr)      0     69  trafic\n",
      "\n",
      "14/24 sujets exploitables, 19 épisodes retenus\n"
     ]
    }
   ],
   "source": [
    "CRITERIA = EpisodeCriteria()\n",
    "\n",
    "reports = {}\n",
    "for label, (entry, series) in SERIES.items():\n",
    "    reports[label] = scan_series(series.filled(), label=label, criteria=CRITERIA)\n",
    "\n",
    "print(f\"{'classe':11s} {'sujet':27s} {'ép.':>4s} {'cand.':>6s}  motif de rejet dominant\")\n",
    "print(\"-\" * 76)\n",
    "for label, (entry, _) in SERIES.items():\n",
    "    report = reports[label]\n",
    "    motif = report.dominant_rejection or \"—\"\n",
    "    print(f\"{entry.category:11s} {entry.label:27s} {len(report.episodes):4d} \"\n",
    "          f\"{report.candidates:6d}  {motif}\")\n",
    "\n",
    "exploitable = sum(report.is_exploitable for report in reports.values())\n",
    "episodes_total = sum(len(report.episodes) for report in reports.values())\n",
    "print(f\"\\n{exploitable}/{len(CORPUS)} sujets exploitables, {episodes_total} épisodes retenus\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c06",
   "metadata": {},
   "source": [
    "### La sélection est biaisée, et il faut le dire\n",
    "\n",
    "Regardons quels sujets ont été écartés : QAnon, désinformation Covid-19, hésitation\n",
    "vaccinale, grand remplacement, Pegasus. Autrement dit, **les cas archétypaux de la théorie.**\n",
    "\n",
    "La raison est structurelle et non accidentelle. Ces sujets ne produisent pas un pic suivi\n",
    "d'une décroissance : leur attention **change de régime et s'installe** sur un palier\n",
    "durable. Or le niveau de fond glissant suit ce palier, et le critère de proéminence n'est\n",
    "alors jamais franchi — la méthode ne les rejette pas, elle **ne les voit pas**.\n",
    "\n",
    "C'est une limite de premier ordre : la procédure sélectionne contre le phénomène même que\n",
    "la théorie cherche à décrire, et pousse la classe « accusation » vers ses épisodes les moins\n",
    "représentatifs. Toute conclusion sur la comparaison entre classes en est affaiblie."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c07",
   "metadata": {},
   "source": [
    "## 4. Un épisode, vu de près\n",
    "\n",
    "Sur une échelle logarithmique, les deux régimes exponentiels apparaissent comme deux\n",
    "segments de droite. C'est ce que les régressions ajustent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c08",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:37:34.480303Z",
     "iopub.status.busy": "2026-08-23T13:37:34.480233Z",
     "iopub.status.idle": "2026-08-23T13:37:34.483491Z",
     "shell.execute_reply": "2026-08-23T13:37:34.483073Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accusation  Pizzagate                   2026-02-04  x33  r_up=+0.842  λ=0.083  γα/λ=11.18\n",
      "discovery   Ondes gravitationnelles     2016-02-12  x125  r_up=+1.195  λ=0.109  γα/λ=12.01\n"
     ]
    }
   ],
   "source": [
    "def episode_frame(label, episode, span=12):\n",
    "    \"\"\"Extrait la fenêtre d'affichage d'un épisode, et les droites ajustées.\"\"\"\n",
    "    entry, series = SERIES[label]\n",
    "    values = series.filled()\n",
    "    left = max(0, episode.peak_index - span)\n",
    "    right = min(len(values), episode.peak_index + 3 * span)\n",
    "    offsets = np.arange(left, right) - episode.peak_index\n",
    "\n",
    "    rise_days = np.arange(-episode.rise.n_points + 1, 1)\n",
    "    decay_days = np.arange(0, episode.decay.n_points)\n",
    "    rise_line = np.exp(episode.rise.intercept + episode.rise.rate * np.arange(episode.rise.n_points))\n",
    "    decay_line = np.exp(episode.decay.intercept + episode.decay.rate * decay_days)\n",
    "\n",
    "    return offsets, values[left:right], rise_days, rise_line, decay_days, decay_line\n",
    "\n",
    "# Deux illustrations, une par classe, choisies comme les épisodes de plus forte amplitude.\n",
    "def strongest(category):\n",
    "    candidates = [\n",
    "        (label, ep)\n",
    "        for label, (entry, _) in SERIES.items()\n",
    "        if entry.category == category\n",
    "        for ep in reports[label].episodes\n",
    "    ]\n",
    "    return max(candidates, key=lambda item: item[1].amplitude)\n",
    "\n",
    "for category in (\"accusation\", \"discovery\"):\n",
    "    label, episode = strongest(category)\n",
    "    day = SERIES[label][1].day(episode.peak_index)\n",
    "    print(f\"{category:11s} {label:27s} {day}  x{episode.amplitude:.0f}  \"\n",
    "          f\"r_up={episode.rise.rate:+.3f}  λ={episode.damping:.3f}  \"\n",
    "          f\"γα/λ={episode.resonance_ratio:.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c09",
   "metadata": {},
   "source": [
    "## 5. L'artefact qu'il fallait corriger\n",
    "\n",
    "Avec des fenêtres délimitées par le retour au niveau de fond, leur **durée varie d'un\n",
    "épisode à l'autre** — de six à quarante-six jours dans ce corpus. Or l'attention ne décroît\n",
    "pas exactement comme une exponentielle : sa queue est plus lourde. Un ajustement\n",
    "exponentiel sur une fenêtre longue capte donc cette queue et produit un $\\lambda$ plus\n",
    "faible.\n",
    "\n",
    "La conséquence est mesurable, et elle est sévère.\n",
    "\n",
    "Un second estimateur, à **horizon fixe**, force les deux fenêtres à la même durée pour tous\n",
    "les épisodes. $\\lambda$ devient alors « le taux d'oubli moyen sur les $H$ premiers jours\n",
    "après le pic » : une grandeur comparable. Le prix est le rejet des épisodes trop brefs pour\n",
    "couvrir l'horizon — d'où la chute de l'effectif quand $H$ augmente."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c10",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:37:34.484167Z",
     "iopub.status.busy": "2026-08-23T13:37:34.484108Z",
     "iopub.status.idle": "2026-08-23T13:37:35.034045Z",
     "shell.execute_reply": "2026-08-23T13:37:35.033559Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Estimateur adaptatif — corrélation de rang avec la durée de la fenêtre :\n",
      "  durée ↔ λ       rho=-0.940  p=0.0000\n",
      "  durée ↔ γα/λ    rho=+0.621  p=0.0045\n",
      "\n",
      "Une corrélation de cette ampleur signifie que λ est déterminé pour l'essentiel\n",
      "par la longueur de la fenêtre, et non par la dynamique du sujet.\n"
     ]
    }
   ],
   "source": [
    "def collect(criteria):\n",
    "    rows = []\n",
    "    for label, (entry, series) in SERIES.items():\n",
    "        for episode in scan_series(series.filled(), label=label, criteria=criteria).episodes:\n",
    "            rows.append({\n",
    "                \"classe\": entry.category,\n",
    "                \"sujet\": entry.label,\n",
    "                \"ratio\": episode.resonance_ratio,\n",
    "                \"r_up\": episode.rise.rate,\n",
    "                \"lambda\": episode.damping,\n",
    "                \"n_decay\": episode.decay.n_points,\n",
    "            })\n",
    "    return rows\n",
    "\n",
    "adaptive = collect(CRITERIA)\n",
    "n_decay = np.array([row[\"n_decay\"] for row in adaptive])\n",
    "lambdas = np.array([row[\"lambda\"] for row in adaptive])\n",
    "ratios = np.array([row[\"ratio\"] for row in adaptive])\n",
    "\n",
    "rho_lambda = stats.spearmanr(n_decay, lambdas)\n",
    "rho_ratio = stats.spearmanr(n_decay, ratios)\n",
    "print(\"Estimateur adaptatif — corrélation de rang avec la durée de la fenêtre :\")\n",
    "print(f\"  durée ↔ λ       rho={rho_lambda.statistic:+.3f}  p={rho_lambda.pvalue:.4f}\")\n",
    "print(f\"  durée ↔ γα/λ    rho={rho_ratio.statistic:+.3f}  p={rho_ratio.pvalue:.4f}\")\n",
    "print(\"\\nUne corrélation de cette ampleur signifie que λ est déterminé pour l'essentiel\")\n",
    "print(\"par la longueur de la fenêtre, et non par la dynamique du sujet.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c11",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:37:35.034877Z",
     "iopub.status.busy": "2026-08-23T13:37:35.034797Z",
     "iopub.status.idle": "2026-08-23T13:37:37.191464Z",
     "shell.execute_reply": "2026-08-23T13:37:37.190936Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "estimateur       n  médiane              IQR          étendue  p (A vs D)\n",
      "--------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "adaptatif       19     4.16 [  3.44,  6.17] [  2.01, 12.01]       0.482\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "horizon 5 j     19     2.52 [  1.92,  3.80] [  1.54,  4.89]       0.129\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "horizon 7 j      9     3.34 [  2.97,  4.21] [  1.81,  6.28]       0.556\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "horizon 10 j     4     3.18 [  2.78,  3.54] [  2.47,  3.78]       0.667\n",
      "\n",
      "γα/λ > 1 dans tous les épisodes, pour tous les estimateurs : True\n"
     ]
    }
   ],
   "source": [
    "ESTIMATORS = {\n",
    "    \"adaptatif\": CRITERIA,\n",
    "    \"horizon 5 j\": EpisodeCriteria(horizon=5),\n",
    "    \"horizon 7 j\": EpisodeCriteria(horizon=7),\n",
    "    \"horizon 10 j\": EpisodeCriteria(horizon=10),\n",
    "}\n",
    "\n",
    "summary = {}\n",
    "print(f\"{'estimateur':14s} {'n':>3s} {'médiane':>8s} {'IQR':>16s} {'étendue':>16s} {'p (A vs D)':>11s}\")\n",
    "print(\"-\" * 74)\n",
    "for name, criteria in ESTIMATORS.items():\n",
    "    rows = collect(criteria)\n",
    "    values = np.array([row[\"ratio\"] for row in rows])\n",
    "    classes = np.array([row[\"classe\"] for row in rows])\n",
    "    accusation = values[classes == \"accusation\"]\n",
    "    discovery = values[classes == \"discovery\"]\n",
    "    p_value = (\n",
    "        stats.mannwhitneyu(accusation, discovery, alternative=\"two-sided\").pvalue\n",
    "        if min(len(accusation), len(discovery)) > 1 else np.nan\n",
    "    )\n",
    "    summary[name] = {\"rows\": rows, \"values\": values, \"classes\": classes, \"p\": p_value}\n",
    "    print(f\"{name:14s} {len(values):3d} {np.median(values):8.2f} \"\n",
    "          f\"[{np.percentile(values, 25):6.2f},{np.percentile(values, 75):6.2f}] \"\n",
    "          f\"[{values.min():6.2f},{values.max():6.2f}] {p_value:11.3f}\")\n",
    "\n",
    "everywhere = all((entry[\"values\"] > 1.0).all() for entry in summary.values())\n",
    "print(f\"\\nγα/λ > 1 dans tous les épisodes, pour tous les estimateurs : {everywhere}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c12",
   "metadata": {},
   "source": [
    "## 6. La prédiction sur la charge émotionnelle n'est pas vérifiée\n",
    "\n",
    "Le modèle fait dépendre l'amplification du produit $\\gamma\\alpha$. À gain algorithmique\n",
    "comparable, un contenu à forte charge émotionnelle devrait donc présenter un rapport plus\n",
    "élevé.\n",
    "\n",
    "Ce n'est pas ce qu'on observe. Aucun estimateur ne produit de différence détectable entre\n",
    "les deux classes, et l'estimation ponctuelle va **dans le sens contraire** à la prédiction :\n",
    "la classe « découverte » présente une médiane légèrement supérieure.\n",
    "\n",
    "Il serait malhonnête d'en tirer une réfutation du mécanisme. Les effectifs sont minuscules\n",
    "(7 à 12 épisodes par classe), le biais de sélection décrit plus haut écarte précisément les\n",
    "cas les plus chargés émotionnellement, et Wikipédia n'est pas le terrain où le mécanisme est\n",
    "censé opérer. Mais il serait tout aussi malhonnête de présenter le mécanisme comme étayé :\n",
    "**il ne l'est pas.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c13",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:37:37.192272Z",
     "iopub.status.busy": "2026-08-23T13:37:37.192205Z",
     "iopub.status.idle": "2026-08-23T13:37:37.636972Z",
     "shell.execute_reply": "2026-08-23T13:37:37.636477Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1380x912 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure, axes = plt.subplots(2, 2, figsize=(11.5, 7.6))\n",
    "example, distribution, sensitivity, artefact = axes.ravel()\n",
    "\n",
    "# (a) Un épisode vu de près, en échelle logarithmique.\n",
    "label, episode = strongest(\"discovery\")\n",
    "offsets, window, rise_days, rise_line, decay_days, decay_line = episode_frame(label, episode)\n",
    "example.semilogy(offsets, np.clip(window, 1, None), \".\", color=PALETTE[\"neutral\"],\n",
    "                 markersize=4, label=\"consultations\")\n",
    "example.semilogy(rise_days, rise_line + episode.baseline, \"-\", color=PALETTE[\"disorder\"],\n",
    "                 label=f\"montée : $r_{{up}}$ = {episode.rise.rate:+.2f}/j\")\n",
    "example.semilogy(decay_days, decay_line + episode.baseline, \"-\", color=PALETTE[\"order\"],\n",
    "                 label=f\"oubli : $\\\\lambda$ = {episode.damping:.2f}/j\")\n",
    "example.axhline(episode.baseline, color=PALETTE[\"neutral\"], linestyle=\":\", linewidth=1.0)\n",
    "example.axvline(0, color=PALETTE[\"field\"], linestyle=\"--\", linewidth=1.0)\n",
    "example.set_xlabel(\"jours autour du pic\")\n",
    "example.set_ylabel(\"consultations / jour\")\n",
    "example.set_title(f\"{label} — $\\\\gamma\\\\alpha/\\\\lambda$ = {episode.resonance_ratio:.1f}\",\n",
    "                  fontsize=10)\n",
    "example.legend(fontsize=7.5, loc=\"lower center\")\n",
    "\n",
    "# (b) Distribution par classe, estimateur à horizon fixe (effectif maximal).\n",
    "chosen = summary[\"horizon 5 j\"]\n",
    "rng = np.random.default_rng(0)\n",
    "for position, (category, colour, name) in enumerate(\n",
    "    [(\"accusation\", PALETTE[\"field\"], \"accusation\"), (\"discovery\", PALETTE[\"remedy\"], \"découverte\")]\n",
    "):\n",
    "    values = chosen[\"values\"][chosen[\"classes\"] == category]\n",
    "    jitter = position + rng.uniform(-0.10, 0.10, size=values.size)\n",
    "    distribution.plot(jitter, values, \"o\", color=colour, markersize=5, alpha=0.8)\n",
    "    distribution.hlines(np.median(values), position - 0.28, position + 0.28,\n",
    "                        color=colour, linewidth=2.4)\n",
    "    distribution.text(position, 1.12, f\"n = {values.size}\",\n",
    "                      ha=\"center\", fontsize=8, color=colour)\n",
    "\n",
    "distribution.axhline(1.0, color=PALETTE[\"neutral\"], linestyle=\"--\", linewidth=1.2)\n",
    "distribution.text(1.45, 1.06, \"seuil $\\\\gamma\\\\alpha = \\\\lambda$\", ha=\"right\",\n",
    "                  fontsize=8, color=PALETTE[\"neutral\"])\n",
    "distribution.set_xticks([0, 1])\n",
    "distribution.set_xticklabels([\"accusation\", \"découverte\"])\n",
    "distribution.set_xlim(-0.5, 1.5)\n",
    "distribution.set_ylabel(\"$\\\\gamma\\\\alpha / \\\\lambda$\")\n",
    "distribution.set_title(f\"Aucun écart détectable (p = {chosen['p']:.2f})\", fontsize=10)\n",
    "\n",
    "# (c) Sensibilité de la médiane au choix d'estimateur.\n",
    "names = list(ESTIMATORS)\n",
    "medians = [np.median(summary[name][\"values\"]) for name in names]\n",
    "lows = [np.percentile(summary[name][\"values\"], 25) for name in names]\n",
    "highs = [np.percentile(summary[name][\"values\"], 75) for name in names]\n",
    "counts = [len(summary[name][\"values\"]) for name in names]\n",
    "\n",
    "positions = np.arange(len(names))\n",
    "sensitivity.errorbar(positions, medians,\n",
    "                     yerr=[np.array(medians) - lows, np.array(highs) - np.array(medians)],\n",
    "                     fmt=\"o\", color=PALETTE[\"order\"], capsize=4, markersize=6)\n",
    "for position, median, count in zip(positions, medians, counts):\n",
    "    sensitivity.annotate(f\"n={count}\", (position, median), textcoords=\"offset points\",\n",
    "                         xytext=(9, 4), fontsize=8, color=PALETTE[\"neutral\"])\n",
    "sensitivity.axhline(1.0, color=PALETTE[\"neutral\"], linestyle=\"--\", linewidth=1.2)\n",
    "sensitivity.set_xticks(positions)\n",
    "sensitivity.set_xticklabels(names, fontsize=8.5)\n",
    "sensitivity.set_ylim(0, max(highs) + 1.0)\n",
    "sensitivity.set_ylabel(\"$\\\\gamma\\\\alpha / \\\\lambda$ (médiane, IQR)\")\n",
    "sensitivity.set_title(\"Robuste en signe, sensible en valeur\", fontsize=10)\n",
    "\n",
    "# (d) L'artefact de l'estimateur adaptatif.\n",
    "artefact.plot(n_decay, lambdas, \"o\", color=PALETTE[\"disorder\"], markersize=5, alpha=0.85)\n",
    "artefact.set_xlabel(\"durée de la fenêtre de décroissance  [jours]\")\n",
    "artefact.set_ylabel(\"$\\\\lambda$ estimé  [/jour]\")\n",
    "artefact.set_title(f\"Artefact adaptatif : rho = {rho_lambda.statistic:+.2f}\", fontsize=10)\n",
    "\n",
    "save_figure(figure, \"fig09_calibration.png\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c14",
   "metadata": {},
   "source": [
    "## 7. Ce que le notebook établit\n",
    "\n",
    "**Le chiffre existe.** Sur 19 épisodes d'attention publics, $\\gamma\\alpha/\\lambda$ vaut\n",
    "entre 1,5 et 12, de médiane 2,5 à 4,2 selon l'estimateur. L'amplification est deux à quatre\n",
    "fois plus rapide que l'oubli. C'est le premier ancrage empirique du modèle, et il donne\n",
    "enfin un ordre de grandeur au paramètre le plus important du mémorandum.\n",
    "\n",
    "**Le critère de signe est vide, et la recommandation doit être réécrite.** Le rapport\n",
    "dépasse 1 dans *tous* les épisodes, sous *tous* les estimateurs — et c'est logique : par\n",
    "construction, un épisode d'attention observable a connu une phase de croissance, donc\n",
    "$r_{\\text{up}} > 0$, donc $\\gamma\\alpha > \\lambda$. Vérifier le signe n'apprend rien.\n",
    "\n",
    "> La recommandation 2 du mémorandum, « interdire les configurations où\n",
    "> $\\gamma\\alpha > \\lambda$ », est **inapplicable telle qu'elle est formulée**. Ce qu'un\n",
    "> régulateur peut contraindre, c'est un **plafond sur le rapport** — et la présente mesure\n",
    "> fournit la référence à partir de laquelle un tel plafond se discute.\n",
    "\n",
    "**La valeur est sensible à la méthode, le signe ne l'est pas.** La médiane varie d'un\n",
    "facteur 1,7 entre estimateurs. Toute valeur citée doit l'être avec son estimateur, et un\n",
    "seuil réglementaire adossé à une valeur unique serait attaquable.\n",
    "\n",
    "**Le mécanisme de la charge émotionnelle n'est pas étayé.** Aucun écart détectable entre les\n",
    "deux classes ($p \\geq 0,13$), et l'estimation ponctuelle va dans le sens contraire à la\n",
    "prédiction.\n",
    "\n",
    "**La méthode est aveugle aux régimes installés.** Les cas archétypaux — QAnon,\n",
    "désinformation sanitaire, hésitation vaccinale — ne produisent pas de pic mais un\n",
    "déplacement durable du niveau d'attention, que la détection par proéminence ne voit pas.\n",
    "C'est la limite la plus lourde, et c'est aussi la piste la plus claire pour la suite :\n",
    "il faut une méthode de détection de **changement de régime**, pas de pic."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c15",
   "metadata": {},
   "source": [
    "## Pistes ouvertes par ce notebook\n",
    "\n",
    "1. **Détection de changement de régime** plutôt que de pic, pour atteindre les dynamiques\n",
    "   d'installation durable — le cas qui compte le plus et qui échappe entièrement à cette\n",
    "   analyse.\n",
    "2. **Résolution infra-quotidienne.** Le motif de rejet dominant est la fenêtre trop courte :\n",
    "   beaucoup d'épisodes montent en un ou deux jours. Des données horaires rendraient\n",
    "   identifiable ce qui ne l'est pas ici.\n",
    "3. **Décroissance non exponentielle.** La queue de l'attention est plus lourde qu'une\n",
    "   exponentielle ; ajuster une loi de puissance ou une somme de deux exponentielles\n",
    "   supprimerait l'artefact de fenêtre au lieu de le contourner par un horizon fixe.\n",
    "4. **Une source dotée d'un algorithme de recommandation**, seule voie pour estimer un\n",
    "   $\\gamma$ de plateforme plutôt qu'un gain d'écosystème. C'est l'objet de l'accès aux\n",
    "   données de l'article 40 du DSA."
   ]
  }
 ],
 "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": "3.12.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}