{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c00",
   "metadata": {},
   "source": [
    "# 11 — Corpus étendu : l'écart de persistance ne se réplique pas\n",
    "\n",
    "Le [notebook 10](10_changement_de_regime.ipynb) a trouvé le premier écart mesuré du projet\n",
    "entre registres émotionnels : les contenus d'accusation élevaient durablement l'attention\n",
    "d'un facteur **×9,2** contre **×2,9** pour les annonces de découverte, à $p = 0{,}08$ sur\n",
    "quatorze observations.\n",
    "\n",
    "Ce notebook le met à l'épreuve sur **440 sujets**. Le résultat est net et il est négatif.\n",
    "\n",
    "**Ce que le corpus étendu établit :**\n",
    "\n",
    "* l'écart de persistance **ne se réplique pas** — ×3,04 contre ×2,90, $p = 0{,}53$. Le\n",
    "  résultat du pilote était un artefact de sélection manuelle ;\n",
    "* un écart apparaît sur le **taux de basculement** — 8,6 % contre 2,7 %, $p = 0{,}014$ —\n",
    "  mais il s'évanouit dès qu'on contrôle le trafic, qui diffère d'un facteur 3,5 entre les\n",
    "  deux registres ;\n",
    "* et l'extension révèle un défaut du nouveau protocole lui-même : l'appartenance à une\n",
    "  catégorie est un **indicateur bruité** du registre émotionnel.\n",
    "\n",
    "Aucun des deux protocoles ne tranche donc la question. C'est cela qu'il fallait apprendre."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c01",
   "metadata": {},
   "source": [
    "## 1. Pourquoi changer de méthode de sélection, et pas seulement de taille\n",
    "\n",
    "Le corpus pilote comptait vingt-quatre sujets choisis à la main. À cette taille, le choix se\n",
    "relit et se conteste. À plusieurs centaines, il ne se relit plus — et c'est précisément là\n",
    "que le biais de sélection devient invisible : rien ne distingue, dans une liste de trois\n",
    "cents titres, ceux qui auraient été retenus pour ce qu'ils montrent.\n",
    "\n",
    "Le corpus étendu remplace donc le choix des **sujets** par le choix des **catégories**. Dix-sept\n",
    "catégories de Wikipédia sont déclarées dans `ide.catalogue.REGISTERS`, et **tout ce qu'elles\n",
    "contiennent** entre dans le pool. L'appartenance d'un article à une catégorie est décidée par\n",
    "les contributeurs de Wikipédia, non par l'auteur de cette analyse.\n",
    "\n",
    "Trois précautions sont prises contre les confondants prévisibles :\n",
    "\n",
    "| Précaution | Motif |\n",
    "|---|---|\n",
    "| des **événements** de part et d'autre | opposer des affaires à des concepts abstraits comparerait des types de sujets, non des registres |\n",
    "| classes **disjointes** | un article des deux registres est écarté, non arbitré |\n",
    "| échantillonnage par **empreinte de titre** | une troncature alphabétique surreprésenterait systématiquement certains sujets |\n",
    "| filtre de **substance** (≥ 10 000 octets) | « Astronomical surveys » à profondeur 1 ramène près de quatre mille entrées de catalogue à trafic nul, qui décimeraient un registre et pas l'autre |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c02",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:04.152630Z",
     "iopub.status.busy": "2026-08-23T13:38:04.152464Z",
     "iopub.status.idle": "2026-08-23T13:38:04.688770Z",
     "shell.execute_reply": "2026-08-23T13:38:04.688222Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "manifeste : catalogue.json   440 sujets\n",
      "filtre de substance : 10000 octets\n",
      "\n",
      "registre      disponibles  substantiels   retenus\n",
      "accusation           3126          1935       220\n",
      "discovery            6322          2705       220\n",
      "\n",
      "articles écartés pour appartenance aux deux registres : 0\n",
      "catégories déclarées : 17\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n",
    "\n",
    "from ide.catalogue import CATALOGUE_PATH, MIN_ARTICLE_BYTES, REGISTERS, load_catalogue\n",
    "from ide.pageviews import load_cached\n",
    "from ide.plotting import PALETTE, save_figure, use_project_style\n",
    "from ide.regime import scan_regime_shifts\n",
    "\n",
    "use_project_style()\n",
    "\n",
    "entries, report = load_catalogue()\n",
    "\n",
    "print(f\"manifeste : {CATALOGUE_PATH.name}   {len(entries)} sujets\")\n",
    "print(f\"filtre de substance : {MIN_ARTICLE_BYTES} octets\\n\")\n",
    "print(f\"{'registre':12s} {'disponibles':>12s} {'substantiels':>13s} {'retenus':>9s}\")\n",
    "for register in (\"accusation\", \"discovery\"):\n",
    "    print(f\"{register:12s} {report['available'][register]:12d} \"\n",
    "          f\"{report['substantial'][register]:13d} {report['kept'][register]:9d}\")\n",
    "print(f\"\\narticles écartés pour appartenance aux deux registres : \"\n",
    "      f\"{len(report['overlap_discarded'])}\")\n",
    "print(f\"catégories déclarées : {len(REGISTERS)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c03",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:04.689524Z",
     "iopub.status.busy": "2026-08-23T13:38:04.689417Z",
     "iopub.status.idle": "2026-08-23T13:38:59.039734Z",
     "shell.execute_reply": "2026-08-23T13:38:59.039321Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sujets analysés      : 440\n",
      "changements détectés : 28\n",
      "dont identifiés      : 1\n"
     ]
    }
   ],
   "source": [
    "profiles, shifts = [], []\n",
    "for entry in entries:\n",
    "    series = load_cached(entry.project, entry.article)\n",
    "    if series is None:\n",
    "        continue\n",
    "    values = np.clip(series.filled(), 1.0, None)\n",
    "    try:\n",
    "        outcome = scan_regime_shifts(values, label=entry.label)\n",
    "    except ValueError:\n",
    "        continue\n",
    "    profiles.append({\n",
    "        \"register\": entry.category,\n",
    "        \"subject\": entry.label,\n",
    "        \"traffic\": float(np.median(values)),\n",
    "        \"shifted\": bool(outcome.shifts),\n",
    "    })\n",
    "    for shift in outcome.shifts:\n",
    "        shifts.append({\n",
    "            \"register\": entry.category,\n",
    "            \"subject\": entry.label,\n",
    "            \"date\": series.day(shift.index),\n",
    "            \"before\": shift.level_before,\n",
    "            \"after\": shift.level_after,\n",
    "            \"lift\": shift.lift,\n",
    "            \"scatter\": shift.fit.scatter,\n",
    "            \"identified\": shift.has_identified_parameters,\n",
    "        })\n",
    "\n",
    "accusation = [p for p in profiles if p[\"register\"] == \"accusation\"]\n",
    "discovery = [p for p in profiles if p[\"register\"] == \"discovery\"]\n",
    "\n",
    "print(f\"sujets analysés      : {len(profiles)}\")\n",
    "print(f\"changements détectés : {len(shifts)}\")\n",
    "print(f\"dont identifiés      : {sum(s['identified'] for s in shifts)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c04",
   "metadata": {},
   "source": [
    "## 2. Un écart apparaît sur le taux de basculement\n",
    "\n",
    "La première mesure semble concluante : les sujets du registre « accusation » basculent trois\n",
    "fois plus souvent que ceux du registre « découverte »."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c05",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.040589Z",
     "iopub.status.busy": "2026-08-23T13:38:59.040507Z",
     "iopub.status.idle": "2026-08-23T13:38:59.043971Z",
     "shell.execute_reply": "2026-08-23T13:38:59.043522Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  accusation  19/220 =   8.6 %\n",
      "  découverte   6/220 =   2.7 %\n",
      "\n",
      "  test exact de Fisher : rapport de cotes = 3.37, p = 0.0120\n"
     ]
    }
   ],
   "source": [
    "def rate(group):\n",
    "    hits = sum(item[\"shifted\"] for item in group)\n",
    "    return hits, len(group), 100.0 * hits / len(group)\n",
    "\n",
    "hits_a, total_a, pct_a = rate(accusation)\n",
    "hits_d, total_d, pct_d = rate(discovery)\n",
    "\n",
    "print(f\"  accusation {hits_a:3d}/{total_a} = {pct_a:5.1f} %\")\n",
    "print(f\"  découverte {hits_d:3d}/{total_d} = {pct_d:5.1f} %\")\n",
    "\n",
    "raw = stats.fisher_exact([[hits_a, total_a - hits_a], [hits_d, total_d - hits_d]])\n",
    "print(f\"\\n  test exact de Fisher : rapport de cotes = {raw.statistic:.2f}, p = {raw.pvalue:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c06",
   "metadata": {},
   "source": [
    "## 3. …mais il est produit par le trafic\n",
    "\n",
    "Avant d'interpréter, il faut vérifier ce qui distingue les deux pools **en dehors** du\n",
    "registre. Et la différence est massive : la détection exige un régime antérieur d'au moins\n",
    "50 consultations par jour, or les deux registres n'ont pas du tout le même niveau\n",
    "d'audience."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c07",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.044679Z",
     "iopub.status.busy": "2026-08-23T13:38:59.044613Z",
     "iopub.status.idle": "2026-08-23T13:38:59.047646Z",
     "shell.execute_reply": "2026-08-23T13:38:59.047266Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  accusation  trafic médian =    39.0 vues/jour   part ≥ 47 vues/jour = 47 %\n",
      "  discovery   trafic médian =    11.0 vues/jour   part ≥ 47 vues/jour = 22 %\n",
      "\n",
      "  Mann-Whitney sur le trafic : p = 4.80e-14\n",
      "\n",
      "  Les sujets d'accusation sont environ trois fois et demie plus consultés.\n",
      "  Le taux de basculement mesuré ci-dessus peut donc n'être qu'un effet d'audience.\n"
     ]
    }
   ],
   "source": [
    "traffic = {\n",
    "    \"accusation\": np.array([p[\"traffic\"] for p in accusation]),\n",
    "    \"discovery\": np.array([p[\"traffic\"] for p in discovery]),\n",
    "}\n",
    "\n",
    "for register, values in traffic.items():\n",
    "    print(f\"  {register:11s} trafic médian = {np.median(values):7.1f} vues/jour   \"\n",
    "          f\"part ≥ 47 vues/jour = {100 * np.mean(values >= 47):.0f} %\")\n",
    "\n",
    "imbalance = stats.mannwhitneyu(traffic[\"accusation\"], traffic[\"discovery\"],\n",
    "                               alternative=\"two-sided\")\n",
    "print(f\"\\n  Mann-Whitney sur le trafic : p = {imbalance.pvalue:.2e}\")\n",
    "print(\"\\n  Les sujets d'accusation sont environ trois fois et demie plus consultés.\")\n",
    "print(\"  Le taux de basculement mesuré ci-dessus peut donc n'être qu'un effet d'audience.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c08",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.048327Z",
     "iopub.status.busy": "2026-08-23T13:38:59.048261Z",
     "iopub.status.idle": "2026-08-23T13:38:59.057550Z",
     "shell.execute_reply": "2026-08-23T13:38:59.057203Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "strate ≥ 47 vues/jour\n",
      "  accusation  17/103 =  16.5 %\n",
      "  découverte   6/48 =  12.5 %\n",
      "  Fisher : rapport de cotes = 1.38, p = 0.631\n",
      "\n",
      "appariement sur le trafic — 173 paires\n",
      "  accusation  12/173 =   6.9 %\n",
      "  découverte   6/173 =   3.5 %\n",
      "  McNemar : 14 paires discordantes, 10 en faveur de l'accusation, p = 0.180\n"
     ]
    }
   ],
   "source": [
    "# --- Contrôle 1 : restriction à la strate de trafic comparable.\n",
    "THRESHOLD = 47.0\n",
    "high_a = [p for p in accusation if p[\"traffic\"] >= THRESHOLD]\n",
    "high_d = [p for p in discovery if p[\"traffic\"] >= THRESHOLD]\n",
    "ha, na, pa_ = rate(high_a)\n",
    "hd, nd, pd_ = rate(high_d)\n",
    "stratified = stats.fisher_exact([[ha, na - ha], [hd, nd - hd]])\n",
    "\n",
    "print(f\"strate ≥ {THRESHOLD:.0f} vues/jour\")\n",
    "print(f\"  accusation {ha:3d}/{na} = {pa_:5.1f} %\")\n",
    "print(f\"  découverte {hd:3d}/{nd} = {pd_:5.1f} %\")\n",
    "print(f\"  Fisher : rapport de cotes = {stratified.statistic:.2f}, p = {stratified.pvalue:.3f}\")\n",
    "\n",
    "# --- Contrôle 2 : appariement de chaque sujet d'accusation à un sujet de découverte\n",
    "# de trafic voisin, à moins d'un facteur deux, sans remise.\n",
    "pool = sorted(discovery, key=lambda item: item[\"traffic\"])\n",
    "pool_logs = np.log([item[\"traffic\"] for item in pool])\n",
    "used, pairs = set(), []\n",
    "for probe in sorted(accusation, key=lambda item: -item[\"traffic\"]):\n",
    "    target = np.log(probe[\"traffic\"])\n",
    "    candidates = [(abs(target - pool_logs[i]), i) for i in range(len(pool)) if i not in used]\n",
    "    if not candidates:\n",
    "        break\n",
    "    distance, index = min(candidates)\n",
    "    if distance <= np.log(2.0):\n",
    "        used.add(index)\n",
    "        pairs.append((probe, pool[index]))\n",
    "\n",
    "matched_a = sum(pair[0][\"shifted\"] for pair in pairs)\n",
    "matched_d = sum(pair[1][\"shifted\"] for pair in pairs)\n",
    "discordant = sum(1 for x, y in pairs if x[\"shifted\"] != y[\"shifted\"])\n",
    "favour = sum(1 for x, y in pairs if x[\"shifted\"] and not y[\"shifted\"])\n",
    "mcnemar = stats.binomtest(favour, discordant, 0.5) if discordant else None\n",
    "\n",
    "print(f\"\\nappariement sur le trafic — {len(pairs)} paires\")\n",
    "print(f\"  accusation {matched_a:3d}/{len(pairs)} = {100 * matched_a / len(pairs):5.1f} %\")\n",
    "print(f\"  découverte {matched_d:3d}/{len(pairs)} = {100 * matched_d / len(pairs):5.1f} %\")\n",
    "if mcnemar is not None:\n",
    "    print(f\"  McNemar : {discordant} paires discordantes, {favour} en faveur de \"\n",
    "          f\"l'accusation, p = {mcnemar.pvalue:.3f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c09",
   "metadata": {},
   "source": [
    "### Lecture\n",
    "\n",
    "Le rapport de cotes brut de 3,4 tombe à 1,38 dans la strate de trafic comparable\n",
    "($p = 0{,}63$), et l'appariement laisse une différence de même sens mais non significative\n",
    "($p = 0{,}18$).\n",
    "\n",
    "**L'écart de taux de basculement est donc pour l'essentiel un effet d'audience, non de\n",
    "registre.** Il subsiste une direction — les sujets d'accusation basculent un peu plus\n",
    "souvent à trafic égal — mais rien qui autorise une conclusion."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c10",
   "metadata": {},
   "source": [
    "## 4. L'écart de persistance ne se réplique pas\n",
    "\n",
    "C'est le test principal, et le résultat est sans ambiguïté."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c11",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.058216Z",
     "iopub.status.busy": "2026-08-23T13:38:59.058157Z",
     "iopub.status.idle": "2026-08-23T13:38:59.061533Z",
     "shell.execute_reply": "2026-08-23T13:38:59.061177Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "corpus                             accusation           découverte        p\n",
      "----------------------------------------------------------------------------\n",
      "pilote (14 sujets)               ×9,20  (n=8)         ×2,90  (n=6)    0.081\n",
      "étendu (440 sujets)             ×3.04  (n=21)         ×2.90  (n=7)    0.533\n",
      "\n",
      "Le facteur trois observé sur le pilote a disparu.\n"
     ]
    }
   ],
   "source": [
    "lifts = {\n",
    "    register: np.array([s[\"lift\"] for s in shifts if s[\"register\"] == register])\n",
    "    for register in (\"accusation\", \"discovery\")\n",
    "}\n",
    "\n",
    "replication = stats.mannwhitneyu(lifts[\"accusation\"], lifts[\"discovery\"],\n",
    "                                 alternative=\"two-sided\")\n",
    "\n",
    "median_a = np.median(lifts[\"accusation\"])\n",
    "median_d = np.median(lifts[\"discovery\"])\n",
    "count_a = lifts[\"accusation\"].size\n",
    "count_d = lifts[\"discovery\"].size\n",
    "\n",
    "print(f\"{'corpus':24s} {'accusation':>20s} {'découverte':>20s} {'p':>8s}\")\n",
    "print(\"-\" * 76)\n",
    "print(f\"{'pilote (14 sujets)':24s} {'×9,20  (n=8)':>20s} {'×2,90  (n=6)':>20s} {0.081:8.3f}\")\n",
    "print(f\"{'étendu (440 sujets)':24s}\"\n",
    "      f\" {f'×{median_a:.2f}  (n={count_a})':>20s}\"\n",
    "      f\" {f'×{median_d:.2f}  (n={count_d})':>20s}\"\n",
    "      f\" {replication.pvalue:8.3f}\")\n",
    "\n",
    "print(\"\\nLe facteur trois observé sur le pilote a disparu.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c12",
   "metadata": {},
   "source": [
    "### Pourquoi le pilote s'était trompé\n",
    "\n",
    "Le corpus pilote contenait les théories du complot les plus connues — QAnon, Pizzagate — et\n",
    "c'est exactement ce qu'une sélection manuelle produit : les cas qui viennent à l'esprit sont\n",
    "les cas extrêmes. Leurs élévations de ×44 et ×18 tiraient la médiane du registre à ×9,2.\n",
    "\n",
    "Le corpus dérivé de catégories contient aussi des dizaines de sujets obscurs du même\n",
    "registre. Sa médiane est ×3,04 — c'est-à-dire indiscernable de celle du registre\n",
    "« découverte ».\n",
    "\n",
    "C'est la démonstration que le changement de protocole de sélection était nécessaire, et\n",
    "qu'il change la réponse."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c13",
   "metadata": {},
   "source": [
    "## 5. Ce que l'extension révèle du nouveau protocole\n",
    "\n",
    "Le corpus étendu corrige un biais et en expose un autre. Regardons les basculements les plus\n",
    "marqués qu'il a retenus dans le registre « accusation »."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c14",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.062152Z",
     "iopub.status.busy": "2026-08-23T13:38:59.062094Z",
     "iopub.status.idle": "2026-08-23T13:38:59.064069Z",
     "shell.execute_reply": "2026-08-23T13:38:59.063752Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  accusation  Lil Tay                                2020-06-27     290 →    2290  ×  7.9\n",
      "  accusation  Watch Dogs (video game)                2020-12-06      57 →     358  ×  6.3\n",
      "  accusation  Million Dollar Extreme                 2016-07-06     111 →     631  ×  5.7\n",
      "  accusation  The Capture (TV series)                2022-07-30     388 →    2133  ×  5.5\n",
      "  discovery   David Baker (biochemist)               2024-09-24      61 →     330  ×  5.4\n",
      "  accusation  Mossack Fonseca                        2019-10-14     286 →    1378  ×  4.8\n",
      "  accusation  Illuminati (game)                      2020-01-16     208 →     737  ×  3.5\n",
      "  accusation  Herbert Kickl                          2024-08-24     144 →     494  ×  3.4\n"
     ]
    }
   ],
   "source": [
    "for record in sorted(shifts, key=lambda item: -item[\"lift\"])[:8]:\n",
    "    print(f\"  {record['register']:11s} {record['subject'][:38]:38s} {record['date']}  \"\n",
    "          f\"{record['before']:6.0f} → {record['after']:7.0f}  ×{record['lift']:5.1f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c15",
   "metadata": {},
   "source": [
    "### Le registre est un indicateur bruité\n",
    "\n",
    "« Lil Tay » figure dans une catégorie de canulars — à cause d'un canular sur sa mort — mais\n",
    "la dynamique d'attention de cet article est celle d'une célébrité, non celle d'une\n",
    "accusation. Il en va de même pour « Watch Dogs (video game) », « Illuminati (game) » ou\n",
    "« The Capture (TV series) » : ces sujets appartiennent aux catégories retenues pour des\n",
    "raisons thématiques, sans que leur audience soit mobilisée par un scandale.\n",
    "\n",
    "C'est une conséquence directe du protocole : **l'appartenance à une catégorie est un\n",
    "indicateur bruité du registre émotionnel.** Et un bruit d'étiquetage ne biaise pas le\n",
    "résultat dans une direction, il **l'attire vers zéro**.\n",
    "\n",
    "Le résultat nul du corpus étendu est donc compatible avec deux lectures :\n",
    "\n",
    "1. il n'existe pas d'écart de persistance entre registres ;\n",
    "2. il en existe un, dilué par l'étiquetage approximatif.\n",
    "\n",
    "Les données présentées ici ne permettent pas de choisir."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c16",
   "metadata": {},
   "source": [
    "## 6. L'identification reste hors de portée\n",
    "\n",
    "Deux ajustements sur vingt-huit ont d'abord été déclarés exploitables, avec des rapports de\n",
    "697 et 5431 — soit des temps d'oubli de plusieurs années, pour une fenêtre d'ajustement de\n",
    "quatre mois. Une transition presque en marche d'escalier n'expose pas le coude qui porte\n",
    "l'information sur $\\lambda$ : l'ajustement épouse la courbe, la dispersion résiduelle est\n",
    "excellente, et le rapport s'envole sans que rien ne le contraigne.\n",
    "\n",
    "Un contrôle d'**observabilité** a donc été ajouté au module : le temps d'oubli doit tenir\n",
    "dans la fenêtre ajustée, faute de quoi les paramètres sont refusés."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c17",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.064698Z",
     "iopub.status.busy": "2026-08-23T13:38:59.064642Z",
     "iopub.status.idle": "2026-08-23T13:38:59.066841Z",
     "shell.execute_reply": "2026-08-23T13:38:59.066416Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "changements retenus                : 28\n",
      "paramètres exploitables            : 1\n",
      "dispersion résiduelle médiane      : 0.55\n",
      "seuil d'acceptation de la dispersion : 0.15\n",
      "\n",
      "Comme sur le corpus pilote, l'identification de γα/λ n'aboutit pas.\n"
     ]
    }
   ],
   "source": [
    "print(f\"changements retenus                : {len(shifts)}\")\n",
    "print(f\"paramètres exploitables            : {sum(s['identified'] for s in shifts)}\")\n",
    "print(f\"dispersion résiduelle médiane      : {np.median([s['scatter'] for s in shifts]):.2f}\")\n",
    "print(f\"seuil d'acceptation de la dispersion : 0.15\")\n",
    "print(\"\\nComme sur le corpus pilote, l'identification de γα/λ n'aboutit pas.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c18",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-23T13:38:59.067457Z",
     "iopub.status.busy": "2026-08-23T13:38:59.067403Z",
     "iopub.status.idle": "2026-08-23T13:38:59.579489Z",
     "shell.execute_reply": "2026-08-23T13:38:59.579102Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1380x936 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure, axes = plt.subplots(2, 2, figsize=(11.5, 7.8))\n",
    "audience, rates, persistence, strata = axes.ravel()\n",
    "\n",
    "# (a) Le confondant : distributions de trafic.\n",
    "bins = np.logspace(0, 4, 30)\n",
    "for register, colour, name in [(\"accusation\", PALETTE[\"field\"], \"accusation\"),\n",
    "                               (\"discovery\", PALETTE[\"remedy\"], \"découverte\")]:\n",
    "    audience.hist(traffic[register], bins=bins, alpha=0.55, color=colour, label=name)\n",
    "audience.axvline(THRESHOLD, color=PALETTE[\"neutral\"], linestyle=\"--\", linewidth=1.3)\n",
    "audience.text(THRESHOLD * 1.15, audience.get_ylim()[1] * 0.75, \"seuil de détection\",\n",
    "              fontsize=7.5, color=PALETTE[\"neutral\"], rotation=90, va=\"top\")\n",
    "audience.set_xscale(\"log\")\n",
    "audience.set_xlabel(\"trafic médian du sujet  [vues/jour]\")\n",
    "audience.set_ylabel(\"nombre de sujets\")\n",
    "audience.set_title(f\"Le confondant : ×3,5 d'écart de trafic (p = {imbalance.pvalue:.0e})\",\n",
    "                   fontsize=10)\n",
    "audience.legend(fontsize=8)\n",
    "\n",
    "# (b) Taux de basculement, brut puis contrôlé.\n",
    "labels = [\"brut\\n(n=440)\", f\"strate ≥{THRESHOLD:.0f}\\n(n={na + nd})\",\n",
    "          f\"apparié\\n(n={2 * len(pairs)})\"]\n",
    "values_a = [pct_a, pa_, 100 * matched_a / len(pairs)]\n",
    "values_d = [pct_d, pd_, 100 * matched_d / len(pairs)]\n",
    "pvalues = [raw.pvalue, stratified.pvalue, mcnemar.pvalue if mcnemar else np.nan]\n",
    "positions = np.arange(3)\n",
    "rates.bar(positions - 0.18, values_a, 0.34, color=PALETTE[\"field\"], label=\"accusation\")\n",
    "rates.bar(positions + 0.18, values_d, 0.34, color=PALETTE[\"remedy\"], label=\"découverte\")\n",
    "for position, pvalue in zip(positions, pvalues):\n",
    "    mark = f\"p = {pvalue:.3f}\" if pvalue >= 0.001 else f\"p = {pvalue:.0e}\"\n",
    "    rates.text(position, max(values_a[position], values_d[position]) + 0.7, mark,\n",
    "               ha=\"center\", fontsize=7.5, color=PALETTE[\"neutral\"])\n",
    "rates.set_xticks(positions)\n",
    "rates.set_xticklabels(labels, fontsize=8)\n",
    "rates.set_ylabel(\"taux de basculement  [%]\")\n",
    "rates.set_title(\"L'écart disparaît quand on contrôle le trafic\", fontsize=10)\n",
    "rates.legend(fontsize=8)\n",
    "\n",
    "# (c) La réplication de l'écart de persistance.\n",
    "pilot = {\"accusation\": [44.2, 18.4, 13.7, 12.1, 6.3, 4.4, 2.8, 2.2],\n",
    "         \"discovery\": [3.3, 3.2, 2.9, 2.9, 2.7, 2.4]}\n",
    "rng = np.random.default_rng(0)\n",
    "for offset, (corpus, source, marker) in enumerate(\n",
    "    [(\"pilote\", pilot, \"s\"), (\"étendu\", {k: list(v) for k, v in lifts.items()}, \"o\")]\n",
    "):\n",
    "    for shift_x, (register, colour) in enumerate(\n",
    "        [(\"accusation\", PALETTE[\"field\"]), (\"discovery\", PALETTE[\"remedy\"])]\n",
    "    ):\n",
    "        values = np.array(source[register], dtype=float)\n",
    "        position = offset * 2.4 + shift_x\n",
    "        jitter = position + rng.uniform(-0.11, 0.11, size=values.size)\n",
    "        persistence.semilogy(jitter, values, marker, color=colour, markersize=5, alpha=0.8)\n",
    "        median = float(np.median(values))\n",
    "        persistence.hlines(median, position - 0.3, position + 0.3, color=colour, linewidth=2.6)\n",
    "        persistence.annotate(f\"×{median:.1f}\", (position, median), textcoords=\"offset points\",\n",
    "                             xytext=(0, 9), ha=\"center\", fontsize=8.5, color=colour,\n",
    "                             fontweight=\"bold\")\n",
    "\n",
    "persistence.set_xticks([0, 1, 2.4, 3.4])\n",
    "persistence.set_xticklabels([\"accusation\", \"découverte\", \"accusation\", \"découverte\"],\n",
    "                            fontsize=7.5, rotation=18, ha=\"right\")\n",
    "persistence.text(0.20, -0.30, \"corpus pilote (14 sujets)\", transform=persistence.transAxes,\n",
    "                 ha=\"center\", fontsize=8.5, color=PALETTE[\"neutral\"])\n",
    "persistence.text(0.80, -0.30, \"corpus étendu (440 sujets)\", transform=persistence.transAxes,\n",
    "                 ha=\"center\", fontsize=8.5, color=PALETTE[\"neutral\"])\n",
    "persistence.set_ylim(1.6, 110.0)\n",
    "persistence.text(0.20, 0.94, \"p = 0,081\", transform=persistence.transAxes, ha=\"center\",\n",
    "                 fontsize=8.5, color=PALETTE[\"neutral\"])\n",
    "persistence.text(0.80, 0.94, f\"p = {replication.pvalue:.2f}\", transform=persistence.transAxes,\n",
    "                 ha=\"center\", fontsize=8.5, color=PALETTE[\"neutral\"])\n",
    "persistence.set_ylabel(\"élévation du palier (×)\")\n",
    "persistence.set_title(\"L'écart du pilote ne se réplique pas\", fontsize=10)\n",
    "\n",
    "# (d) Taux de basculement par strate de trafic.\n",
    "all_traffic = np.array([p[\"traffic\"] for p in profiles])\n",
    "edges = np.percentile(all_traffic, [0, 33, 66, 100])\n",
    "centres, series_a, series_d = [], [], []\n",
    "for low, high in zip(edges[:-1], edges[1:]):\n",
    "    centres.append(np.sqrt(max(low, 1.0) * high))\n",
    "    for group, target in ((accusation, series_a), (discovery, series_d)):\n",
    "        inside = [p for p in group if low <= p[\"traffic\"] <= high]\n",
    "        target.append(100 * sum(p[\"shifted\"] for p in inside) / max(len(inside), 1))\n",
    "\n",
    "strata.semilogx(centres, series_a, \"o-\", color=PALETTE[\"field\"], markersize=6,\n",
    "                label=\"accusation\")\n",
    "strata.semilogx(centres, series_d, \"s-\", color=PALETTE[\"remedy\"], markersize=6,\n",
    "                label=\"découverte\")\n",
    "strata.set_xlabel(\"trafic médian de la strate  [vues/jour]\")\n",
    "strata.set_ylabel(\"taux de basculement  [%]\")\n",
    "strata.set_title(\"Le basculement suit l'audience, non le registre\", fontsize=10)\n",
    "strata.legend(fontsize=8)\n",
    "\n",
    "save_figure(figure, \"fig11_corpus_etendu.png\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c19",
   "metadata": {},
   "source": [
    "## 7. Ce que le notebook établit\n",
    "\n",
    "**L'écart de persistance du corpus pilote était un artefact de sélection.** ×9,2 contre ×2,9\n",
    "sur quatorze sujets choisis à la main devient ×3,04 contre ×2,90 sur 440 sujets dérivés de\n",
    "catégories ($p = 0{,}53$). Le protocole de sélection ne modifiait pas la précision du\n",
    "résultat : il en modifiait le signe.\n",
    "\n",
    "**L'écart de taux de basculement est un effet d'audience.** Le rapport de cotes brut de 3,4\n",
    "($p = 0{,}014$) tombe à 1,38 dans une strate de trafic comparable ($p = 0{,}63$). Les sujets\n",
    "d'accusation sont trois fois et demie plus consultés que ceux de découverte, et la détection\n",
    "exige un seuil d'audience : le registre n'explique pas ce que l'audience explique déjà.\n",
    "\n",
    "**Le nouveau protocole a son propre défaut.** L'appartenance à une catégorie est un\n",
    "indicateur bruité du registre — « Lil Tay » est dans une catégorie de canulars, mais son\n",
    "audience est celle d'une célébrité. Un bruit d'étiquetage attire tout écart vers zéro : le\n",
    "résultat nul est donc compatible avec l'absence d'effet **comme** avec un effet dilué.\n",
    "\n",
    "**Aucun des deux protocoles ne tranche.** Le pilote était biaisé par la sélection, l'étendu\n",
    "est bruité par l'étiquetage. Ce n'est pas une impasse mais une spécification : un troisième\n",
    "protocole devrait combiner un pool dérivé de catégories — pour l'absence de biais de\n",
    "sélection — et une **validation du registre sujet par sujet**, indépendante de l'issue\n",
    "mesurée. C'est un travail d'annotation, non de calcul.\n",
    "\n",
    "**Et il faut le dire pour ce que c'est :** la seule différence entre registres émotionnels que\n",
    "le projet avait mesurée ne survit pas à sa vérification. Le mécanisme de la charge émotionnelle\n",
    "$\\alpha$ reste, à ce stade, sans appui empirique — ni par le taux d'amplification\n",
    "([notebook 09](09_calibration_visibilite.ipynb)), ni par la persistance."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c20",
   "metadata": {},
   "source": [
    "## Pistes ouvertes\n",
    "\n",
    "1. **Annoter le registre à la main sur le pool dérivé de catégories.** C'est le seul moyen\n",
    "   de séparer l'absence d'effet de sa dilution. Deux cents sujets annotés en aveugle — sans\n",
    "   voir leur série — suffiraient.\n",
    "2. **Apparier sur le trafic dès la construction du corpus**, plutôt qu'en aval : tirer les\n",
    "   sujets par paires de trafic comparable rendrait la comparaison immunisée au confondant\n",
    "   principal.\n",
    "3. **Abaisser le seuil de détection en agrégeant par semaine.** Le taux de basculement de\n",
    "   2 à 8 % laisse plus de neuf sujets sur dix sans mesure ; une résolution hebdomadaire\n",
    "   ferait passer le seuil d'audience et multiplierait les observations.\n",
    "4. **Tester la spécificité du détecteur sur un registre témoin** — des sujets sans charge\n",
    "   émotionnelle particulière et à trafic apparié — pour établir un taux de basculement de\n",
    "   référence."
   ]
  }
 ],
 "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
}