added tag hieararchy

jupyter_notebooks/Section_1_Figure_1_image_grid.ipynb

@@ -214,7 +214,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "id": "88a6ca8e-51ee-4d27-ba0c-5f026d5750df",
    "metadata": {
     "execution": {
@@ -225,14 +225,23 @@
      "shell.execute_reply.started": "2025-02-08T21:58:52.986878Z"
     }
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Scanning directory: /home/lauhp/000_PHD/000_010_PUBLICATION/CODE/pm-paper/data/sorted/image_metadata\n",
+      "No JSON files found in the directory.\n"
+     ]
+    }
+   ],
    "source": [
-    "#download_and_save_data(input_dir, images)  "
+    "download_and_save_data(input_dir, images)  "
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "id": "f247091b-7b45-4cd4-a6ca-3d4dea414e19",
    "metadata": {
     "execution": {
@@ -253,6 +262,13 @@
     }
    ],
    "source": [
+    "import random\n",
+    "\n",
+    "# For creating the figure grid\n",
+    "images = []\n",
+    "all_valid_images = []\n",
+    "\n",
+    "\n",
     "directory = current_dir.parent / 'data/sorted/images/'\n",
     "#directory = '/home/lauwag/shares/laura_wagner/Civitai_page_analysis/Civitai_dataset/dataset/chronological/full/prompts-images/2024/2024-05/2024-05-31/'\n",
     "output = current_dir.parent / 'plots/grid10x120.png'\n",
@@ -331,19 +347,31 @@
     "\n",
     "file_types = ('png', 'jpg', 'jpeg')  # Define acceptable image file types\n",
     "\n",
-    "images = []\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "# First, collect all valid image paths\n",
     "for root, dirs, files in os.walk(directory):\n",
     "    for file in files:\n",
     "        if file.lower().endswith(file_types):\n",
     "            image_path = os.path.join(root, file)\n",
     "            json_path = image_path.rsplit('.', 1)[0] + '.json'\n",
     "            if os.path.exists(json_path):\n",
-    "                img = process_image(image_path, json_path, cell_size)\n",
-    "                images.append(img)\n",
-    "            if len(images) == grid_size[0] * grid_size[1]:\n",
-    "                break\n",
-    "    if len(images) == grid_size[0] * grid_size[1]:\n",
-    "        break\n",
+    "                all_valid_images.append((image_path, json_path))\n",
+    "\n",
+    "# Randomly sample from valid images\n",
+    "num_needed = grid_size[0] * grid_size[1]\n",
+    "if len(all_valid_images) >= num_needed:\n",
+    "    random.seed(42)  # For reproducibility\n",
+    "    sampled_images = random.sample(all_valid_images, num_needed)\n",
+    "    \n",
+    "    for image_path, json_path in sampled_images:\n",
+    "        img = process_image(image_path, json_path, cell_size)\n",
+    "        images.append(img)\n",
+    "else:\n",
+    "    print(f\"Warning: Only {len(all_valid_images)} valid images found, need {num_needed}\")\n",
+    "\n",
     "\n",
     "# Create the grid image\n",
     "grid_img = Image.new('RGB', (grid_size[1] * cell_size, grid_size[0] * cell_size))\n",

jupyter_notebooks/Section_2-3-4_Figure_8_Step_1_LLM_annotation.ipynb

@@ -1919,9 +1919,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "pm-paper",
+   "display_name": "latm",
    "language": "python",
-   "name": "pm-paper"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -1933,7 +1933,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.13"
+   "version": "3.10.15"
   }
  },
  "nbformat": 4,

jupyter_notebooks/Section_2-3-4_Figure_8_Step_2_response_comparison_and_consensus_extraction.ipynb

@@ -2831,9 +2831,258 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 5,
    "id": "70403f7c-6ea9-4f21-9704-aec0c37a591b",
    "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Original rows: 21400\n",
+      "Rows with name consensus: 16464\n",
+      "Unique persons after aggregation: 8242\n",
+      "Saved to: /home/lauhp/000_PHD/000_010_PUBLICATION/CODE/pm-paper/data/CSV/final_aggregated.csv\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>name</th>\n",
+       "      <th>model_ids</th>\n",
+       "      <th>number_of_models</th>\n",
+       "      <th>consensus_profession</th>\n",
+       "      <th>consensus_gender</th>\n",
+       "      <th>consensus_country</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>AJ Applegate</td>\n",
+       "      <td>[1335260]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>adult performer</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>USA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>AJ Langer</td>\n",
+       "      <td>[1216782]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>actor</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>USA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Aaliyah Dana Haughton</td>\n",
+       "      <td>[221782, 230769, 1191059, 1564721]</td>\n",
+       "      <td>4</td>\n",
+       "      <td>singer/musician</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>USA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Aaliyah Love</td>\n",
+       "      <td>[305961]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>adult performer</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>USA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Aaron Boone</td>\n",
+       "      <td>[437968]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>sports professional</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>USA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8237</th>\n",
+       "      <td>Özge Gürel</td>\n",
+       "      <td>[144371]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>actor</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>Türkiye</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8238</th>\n",
+       "      <td>Özge Özacar</td>\n",
+       "      <td>[975857]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>actor</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>Türkiye</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8239</th>\n",
+       "      <td>Özge Özberk</td>\n",
+       "      <td>[199114]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>actor</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>Türkiye</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8240</th>\n",
+       "      <td>Şener Şen</td>\n",
+       "      <td>[125489]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>actor</td>\n",
+       "      <td>Male</td>\n",
+       "      <td>Türkiye</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8241</th>\n",
+       "      <td>Şifanur Gül</td>\n",
+       "      <td>[133589]</td>\n",
+       "      <td>1</td>\n",
+       "      <td>model</td>\n",
+       "      <td>Female</td>\n",
+       "      <td>Türkiye</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>8242 rows × 6 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                       name                           model_ids  \\\n",
+       "0              AJ Applegate                           [1335260]   \n",
+       "1                 AJ Langer                           [1216782]   \n",
+       "2     Aaliyah Dana Haughton  [221782, 230769, 1191059, 1564721]   \n",
+       "3              Aaliyah Love                            [305961]   \n",
+       "4               Aaron Boone                            [437968]   \n",
+       "...                     ...                                 ...   \n",
+       "8237             Özge Gürel                            [144371]   \n",
+       "8238            Özge Özacar                            [975857]   \n",
+       "8239            Özge Özberk                            [199114]   \n",
+       "8240              Şener Şen                            [125489]   \n",
+       "8241            Şifanur Gül                            [133589]   \n",
+       "\n",
+       "      number_of_models consensus_profession consensus_gender consensus_country  \n",
+       "0                    1      adult performer           Female               USA  \n",
+       "1                    1                actor           Female               USA  \n",
+       "2                    4      singer/musician           Female               USA  \n",
+       "3                    1      adult performer           Female               USA  \n",
+       "4                    1  sports professional             Male               USA  \n",
+       "...                ...                  ...              ...               ...  \n",
+       "8237                 1                actor           Female           Türkiye  \n",
+       "8238                 1                actor           Female           Türkiye  \n",
+       "8239                 1                actor           Female           Türkiye  \n",
+       "8240                 1                actor             Male           Türkiye  \n",
+       "8241                 1                model           Female           Türkiye  \n",
+       "\n",
+       "[8242 rows x 6 columns]"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "from pathlib import Path\n",
+    "from collections import Counter\n",
+    "\n",
+    "current_dir = Path.cwd()\n",
+    "input_file = current_dir.parent / \"data/CSV/analyzed_llm_agreement_consensus_va.csv\"\n",
+    "output_file = current_dir.parent / \"data/CSV/final_aggregated.csv\"\n",
+    "\n",
+    "# Load the data\n",
+    "df = pd.read_csv(input_file)\n",
+    "\n",
+    "# Function to get consensus name from the three LLM columns (at least 2 must agree)\n",
+    "def get_consensus_name(row):\n",
+    "    names = [row['gemma_full_name'], row['mistral_full_name'], row['qwen_full_name']]\n",
+    "    # Filter out NaN/None values\n",
+    "    names = [n for n in names if pd.notna(n)]\n",
+    "    \n",
+    "    if len(names) == 0:\n",
+    "        return None\n",
+    "    \n",
+    "    # Count occurrences of each name\n",
+    "    name_counts = Counter(names)\n",
+    "    \n",
+    "    # Find names with at least 2 occurrences\n",
+    "    for name, count in name_counts.most_common():\n",
+    "        if count >= 2:\n",
+    "            return name\n",
+    "    \n",
+    "    # If no consensus (all 3 different), return None or first non-null\n",
+    "    return None\n",
+    "\n",
+    "# Create consensus_name column\n",
+    "df['consensus_name'] = df.apply(get_consensus_name, axis=1)\n",
+    "\n",
+    "# Filter out rows without consensus name\n",
+    "df_with_consensus = df[df['consensus_name'].notna()].copy()\n",
+    "\n",
+    "# Aggregate by consensus_name\n",
+    "aggregated = df_with_consensus.groupby('consensus_name').agg(\n",
+    "    model_ids=('id', lambda x: list(x)),\n",
+    "    number_of_models=('id', 'count'),\n",
+    "    consensus_profession=('consensus_primary_profession', 'first'),\n",
+    "    consensus_gender=('consensus_gender', 'first'),\n",
+    "    consensus_country=('consensus_country', 'first')\n",
+    ").reset_index()\n",
+    "\n",
+    "# Rename consensus_name to name for clarity\n",
+    "aggregated = aggregated.rename(columns={'consensus_name': 'name'})\n",
+    "\n",
+    "# Reorder columns\n",
+    "aggregated = aggregated[['name', 'model_ids', 'number_of_models', 'consensus_profession', 'consensus_gender', 'consensus_country']]\n",
+    "\n",
+    "# Save to CSV\n",
+    "aggregated.to_csv(output_file, index=False)\n",
+    "\n",
+    "print(f\"Original rows: {len(df)}\")\n",
+    "print(f\"Rows with name consensus: {len(df_with_consensus)}\")\n",
+    "print(f\"Unique persons after aggregation: {len(aggregated)}\")\n",
+    "print(f\"Saved to: {output_file}\")\n",
+    "\n",
+    "aggregated"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d1cd2440",
+   "metadata": {},
    "outputs": [],
    "source": []
   }

jupyter_notebooks/Section_2-3-4__Figure_8a_sunburst_gender.ipynb

@@ -9,14 +9,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "✓ Saved 8a.json\n"
+      "✓ Loaded 8242 records\n",
+      "✓ Saved sunburst_gender_A.json\n"
      ]
     }
    ],
@@ -24,77 +25,118 @@
     "import pandas as pd\n",
     "from collections import defaultdict\n",
     "import json\n",
-    "from pathlib import Path \n",
+    "from pathlib import Path\n",
     "\n",
     "current_dir = Path.cwd()\n",
-    "sunburst_json = current_dir.parent / \"public/json/8a.json\"\n",
+    "sunburst_path = current_dir.parent / \"public/json/sunburst_gender_A.json\"\n",
     "\n",
     "# Load consensus CSV\n",
-    "consensus_file = current_dir.parent / \"data/CSV/analyzed_llm_agreement_consensus.csv\"\n",
+    "consensus_file = current_dir.parent / \"data/CSV/final_aggregated.csv\"\n",
     "df = pd.read_csv(consensus_file)\n",
     "\n",
-    "# ---- Normalize Gender (group Non-binary and Unknown into 'Other') ----\n",
+    "print(f\"✓ Loaded {len(df)} records\")\n",
+    "\n",
+    "# ============================================================\n",
+    "# 1. NORMALIZE GENDER\n",
+    "# ============================================================\n",
     "def normalize_gender(g):\n",
-    "    g = str(g).strip().lower()\n",
-    "    if g in [\"female\", \"woman\", \"female (group)\", \"female (transgender)\", \"female (virtual persona)\", \"female (group members)\"]:\n",
+    "    if not isinstance(g, str):\n",
+    "        g = str(g)\n",
+    "    g = g.strip().lower()\n",
+    "    \n",
+    "    if g in [\"female\", \"woman\", \"female (group)\", \"female (transgender)\", \n",
+    "             \"female (virtual persona)\", \"female (group members)\"]:\n",
     "        return \"Female\"\n",
     "    elif g in [\"male\", \"male (android)\", \"male (character)\"]:\n",
     "        return \"Male\"\n",
     "    else:\n",
     "        return \"Other\"\n",
     "\n",
-    "df['gender_normalized'] = df['consensus_gender'].apply(normalize_gender)\n",
+    "df[\"gender_normalized\"] = df[\"consensus_gender\"].apply(normalize_gender)\n",
     "\n",
-    "# ---- Step 1: Limit to top 10 countries ----\n",
-    "df['country_cleaned'] = df['consensus_country'].apply(lambda x: x if x not in ['Unknown', '', None] else 'Other')\n",
-    "top_countries = df['country_cleaned'].value_counts().nlargest(12).index.tolist()\n",
-    "df['country_limited'] = df['country_cleaned'].apply(lambda x: x if x in top_countries else 'Other')\n",
+    "# ============================================================\n",
+    "# 2. NORMALIZE PROFESSIONS\n",
+    "# ============================================================\n",
+    "def normalize_profession(x: str):\n",
+    "    if not isinstance(x, str) or x.strip() == \"\" or x.lower() == \"unknown\":\n",
+    "        return \"Other\"\n",
+    "    x = x.strip().lower()\n",
+    "    mapping = {\n",
+    "        \"actor\": \"Actor\",\n",
+    "        \"model\": \"Model\",\n",
+    "        \"adult performer\": \"Adult Performer\",\n",
+    "        \"singer/musician\": \"Singer, Musician\",\n",
+    "        \"online personality\": \"Online Personality\",\n",
+    "        \"sports professional\": \"Sports Professional\",\n",
+    "        \"voice actor/asmr\": \"Voice Actor\",\n",
+    "        \"public figure\": \"Public Figure\",\n",
+    "        \"tv personality\": \"Other\",\n",
+    "    }\n",
+    "    return mapping.get(x, \"Other\")\n",
     "\n",
-    "# ---- Step 2: Normalize and limit professions ----\n",
-    "valid_categories = [\n",
-    "    \"actor\", \"adult performer\", \"singer/musician\", \"model\",\n",
-    "    \"online personality\", \"tv personality\", \"voice actor/asmr\", \"public figure\", \"sports professional\"\n",
-    "]\n",
+    "df[\"profession_clean\"] = df[\"consensus_profession\"].apply(normalize_profession)\n",
     "\n",
-    "def remap_profession(profession):\n",
-    "    profession_lower = str(profession).strip().lower()\n",
-    "    if profession_lower == 'unknown' or profession_lower not in valid_categories:\n",
-    "        return 'Other'\n",
-    "    elif profession_lower == 'fictional character':\n",
-    "        return 'actor'\n",
-    "    elif profession_lower in ['voice actor', 'voice actor/asmr']:\n",
-    "        return 'voice actor/ASMR'\n",
-    "    return profession_lower\n",
+    "# Define top professions to keep\n",
+    "top_prof = [\n",
+    "    \"Actor\",\n",
+    "    \"Model\", \n",
+    "    \"Adult Performer\",\n",
+    "    \"Singer, Musician\",\n",
+    "    \"Online Personality\",\n",
+    "    \"Sports Professional\",\n",
+    "    \"Voice Actor\",\n",
+    "    \"Public Figure\",\n",
+    "    \"Other\"\n",
+    "]\n",
     "\n",
-    "df['profession_limited'] = df['consensus_primary_profession'].apply(remap_profession)\n",
+    "# Re-limit professions, everything else → Other\n",
+    "df[\"profession_limited\"] = df[\"profession_clean\"].apply(\n",
+    "    lambda x: x if x in top_prof else \"Other\"\n",
+    ")\n",
     "\n",
-    "# ---- Step 3: Group by gender and profession ----\n",
-    "sunburst_data = df.groupby(['gender_normalized', 'profession_limited']).size().reset_index(name='count')\n",
+    "# ============================================================\n",
+    "# 3. GROUP INTO SUNBURST STRUCTURE\n",
+    "# ============================================================\n",
+    "sunburst_data = (\n",
+    "    df.groupby([\"gender_normalized\", \"profession_limited\"])\n",
+    "        .size()\n",
+    "        .reset_index(name=\"count\")\n",
+    ")\n",
     "\n",
-    "# ---- Step 4: Create nested structure for D3.js ----\n",
     "sunburst_dict = {\"name\": \"root\", \"children\": []}\n",
     "gender_map = defaultdict(list)\n",
     "\n",
     "for _, row in sunburst_data.iterrows():\n",
-    "    gender = row['gender_normalized']\n",
-    "    profession = row['profession_limited']\n",
-    "    count = int(row['count'])\n",
-    "    gender_map[gender].append({\"name\": profession, \"value\": count})\n",
-    "\n",
-    "# Sort 'Other' professions to appear last\n",
-    "for gender, professions in gender_map.items():\n",
-    "    professions_sorted = sorted(professions, key=lambda d: (d[\"name\"] == \"Other\", d[\"name\"]))\n",
-    "    gender_map[gender] = professions_sorted\n",
-    "\n",
-    "# Build the final nested structure\n",
-    "for gender, professions in gender_map.items():\n",
-    "    sunburst_dict[\"children\"].append({\"name\": gender, \"children\": professions})\n",
-    "\n",
-    "# ---- Step 5: Save to a JSON file ----\n",
-    "with open(sunburst_json, \"w\", encoding='utf-8') as f:\n",
+    "    g = row[\"gender_normalized\"]\n",
+    "    p = row[\"profession_limited\"]\n",
+    "    v = int(row[\"count\"])\n",
+    "    gender_map[g].append({\"name\": p, \"value\": v})\n",
+    "\n",
+    "# Sort professions inside each gender so \"Other\" is last\n",
+    "for g, profs in gender_map.items():\n",
+    "    profs_sorted = sorted(profs, key=lambda d: (d[\"name\"] == \"Other\", d[\"name\"]))\n",
+    "    gender_map[g] = profs_sorted\n",
+    "\n",
+    "# Calculate total datapoints per gender and sort\n",
+    "gender_totals = []\n",
+    "for g, profs in gender_map.items():\n",
+    "    total = sum(p[\"value\"] for p in profs)\n",
+    "    gender_totals.append((g, total, profs))\n",
+    "\n",
+    "# Sort by total (descending), but put \"Other\" last\n",
+    "gender_totals.sort(key=lambda x: (x[0] == \"Other\", -x[1]))\n",
+    "\n",
+    "# Build final JSON with sorted genders\n",
+    "for g, total, profs in gender_totals:\n",
+    "    sunburst_dict[\"children\"].append({\"name\": g, \"children\": profs})\n",
+    "\n",
+    "# ============================================================\n",
+    "# 4. SAVE JSON\n",
+    "# ============================================================\n",
+    "with open(sunburst_path, \"w\", encoding=\"utf-8\") as f:\n",
     "    json.dump(sunburst_dict, f, ensure_ascii=False, indent=2)\n",
     "\n",
-    "print(\"✓ Saved 8a.json\")\n"
+    "print(\"✓ Saved sunburst_gender_A.json\")"
    ]
   },
   {

jupyter_notebooks/Section_2-3-4__Figure_8b_sunburst_profession.ipynb

@@ -9,7 +9,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -31,7 +31,7 @@
     "sunburst_path = current_dir.parent / \"public/json/sunburst_countries_A.json\"\n",
     "\n",
     "# Load consensus CSV\n",
-    "consensus_file = current_dir.parent / \"data/CSV/analyzed_llm_agreement_consensus_va.csv\"\n",
+    "consensus_file = current_dir.parent / \"data/CSV/final_aggregated.csv\"\n",
     "df = pd.read_csv(consensus_file)\n",
     "\n",
     "# ============================================================\n",
@@ -84,7 +84,7 @@
     "    }\n",
     "    return mapping.get(x, \"Other\")\n",
     "\n",
-    "df[\"profession_clean\"] = df[\"consensus_primary_profession\"].apply(normalize_profession)\n",
+    "df[\"profession_clean\"] = df[\"consensus_profession\"].apply(normalize_profession)\n",
     "\n",
     "\n",
     "df[\"profession_limited\"] = df[\"profession_clean\"]\n",
@@ -147,15 +147,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "✓ Filtered to 16059 records published on or before December 31, 2024\n",
-      "✓ Saved sunburst_countries_A.json (2024 data only)\n"
+     "ename": "KeyError",
+     "evalue": "'publishedAt'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
+      "File \u001b[0;32m~/anaconda3/envs/latm/lib/python3.10/site-packages/pandas/core/indexes/base.py:3805\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   3804\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 3805\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcasted_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3806\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
+      "File \u001b[0;32mindex.pyx:167\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
+      "File \u001b[0;32mindex.pyx:196\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
+      "File \u001b[0;32mpandas/_libs/hashtable_class_helper.pxi:7081\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
+      "File \u001b[0;32mpandas/_libs/hashtable_class_helper.pxi:7089\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
+      "\u001b[0;31mKeyError\u001b[0m: 'publishedAt'",
+      "\nThe above exception was the direct cause of the following exception:\n",
+      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[2], line 17\u001b[0m\n\u001b[1;32m     11\u001b[0m df \u001b[38;5;241m=\u001b[39m pd\u001b[38;5;241m.\u001b[39mread_csv(consensus_file)\n\u001b[1;32m     13\u001b[0m \u001b[38;5;66;03m# ============================================================\u001b[39;00m\n\u001b[1;32m     14\u001b[0m \u001b[38;5;66;03m# FILTER DATA UP TO DECEMBER 31, 2024\u001b[39;00m\n\u001b[1;32m     15\u001b[0m \u001b[38;5;66;03m# ============================================================\u001b[39;00m\n\u001b[1;32m     16\u001b[0m \u001b[38;5;66;03m# Convert publishedAt to datetime\u001b[39;00m\n\u001b[0;32m---> 17\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpublishedAt\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m pd\u001b[38;5;241m.\u001b[39mto_datetime(\u001b[43mdf\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mpublishedAt\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m]\u001b[49m, errors\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcoerce\u001b[39m\u001b[38;5;124m\"\u001b[39m, utc\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m     19\u001b[0m \u001b[38;5;66;03m# Filter to only include data up to December 31, 2024\u001b[39;00m\n\u001b[1;32m     20\u001b[0m \u001b[38;5;66;03m# Make cutoff_date timezone-aware (UTC) to match publishedAt\u001b[39;00m\n\u001b[1;32m     21\u001b[0m cutoff_date \u001b[38;5;241m=\u001b[39m pd\u001b[38;5;241m.\u001b[39mTimestamp(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m2024-12-31 23:59:59\u001b[39m\u001b[38;5;124m\"\u001b[39m, tz\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mUTC\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
+      "File \u001b[0;32m~/anaconda3/envs/latm/lib/python3.10/site-packages/pandas/core/frame.py:4102\u001b[0m, in \u001b[0;36mDataFrame.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   4100\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcolumns\u001b[38;5;241m.\u001b[39mnlevels \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[1;32m   4101\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_getitem_multilevel(key)\n\u001b[0;32m-> 4102\u001b[0m indexer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcolumns\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4103\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m is_integer(indexer):\n\u001b[1;32m   4104\u001b[0m     indexer \u001b[38;5;241m=\u001b[39m [indexer]\n",
+      "File \u001b[0;32m~/anaconda3/envs/latm/lib/python3.10/site-packages/pandas/core/indexes/base.py:3812\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   3807\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(casted_key, \u001b[38;5;28mslice\u001b[39m) \u001b[38;5;129;01mor\u001b[39;00m (\n\u001b[1;32m   3808\u001b[0m         \u001b[38;5;28misinstance\u001b[39m(casted_key, abc\u001b[38;5;241m.\u001b[39mIterable)\n\u001b[1;32m   3809\u001b[0m         \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28many\u001b[39m(\u001b[38;5;28misinstance\u001b[39m(x, \u001b[38;5;28mslice\u001b[39m) \u001b[38;5;28;01mfor\u001b[39;00m x \u001b[38;5;129;01min\u001b[39;00m casted_key)\n\u001b[1;32m   3810\u001b[0m     ):\n\u001b[1;32m   3811\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m InvalidIndexError(key)\n\u001b[0;32m-> 3812\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m   3813\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m:\n\u001b[1;32m   3814\u001b[0m     \u001b[38;5;66;03m# If we have a listlike key, _check_indexing_error will raise\u001b[39;00m\n\u001b[1;32m   3815\u001b[0m     \u001b[38;5;66;03m#  InvalidIndexError. Otherwise we fall through and re-raise\u001b[39;00m\n\u001b[1;32m   3816\u001b[0m     \u001b[38;5;66;03m#  the TypeError.\u001b[39;00m\n\u001b[1;32m   3817\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_indexing_error(key)\n",
+      "\u001b[0;31mKeyError\u001b[0m: 'publishedAt'"
      ]
     }
    ],
@@ -169,7 +182,7 @@
     "sunburst_path = current_dir.parent / \"public/json/sunburst_countries_A.json\"\n",
     "\n",
     "# Load consensus CSV\n",
-    "consensus_file = current_dir.parent / \"data/CSV/analyzed_llm_agreement_consensus_va.csv\"\n",
+    "consensus_file = current_dir.parent / \"data/CSV/final_aggregated.csv\"\n",
     "df = pd.read_csv(consensus_file)\n",
     "\n",
     "# ============================================================\n",

public/Figure_13_interactive.html

@@ -0,0 +1,403 @@
+<!DOCTYPE html>
+<html lang="en" class="light-mode">
+<head>
+  <meta charset="UTF-8">
+  <title>Danbooru Tag Taxonomy</title>
+  <script src="https://d3js.org/d3.v7.min.js"></script>
+  <link rel="stylesheet" href="styles.css">
+  <style>
+    /* Tree-specific styles (not in shared styles.css) */
+    html, body { overflow: auto; }
+    svg { position: static; margin-top: 0; z-index: auto; }
+    .tree-svg {
+      width: 100%;
+      overflow: visible;
+      display: block;
+    }
+    .link {
+      fill: none;
+      stroke: #ccc;
+      stroke-width: 1.5px;
+    }
+    .node circle {
+      stroke: #555;
+      stroke-width: 1.5px;
+      cursor: pointer;
+    }
+    .node.leaf circle {
+      cursor: default;
+    }
+    .node text {
+      font-size: 12px;
+      fill: #333;
+      pointer-events: none;
+    }
+    .node.leaf text {
+      font-size: 10px;
+      fill: #666;
+    }
+    .node:hover circle {
+      stroke: #000;
+      stroke-width: 2.5px;
+    }
+    .tree-controls {
+      margin: 0.5em 0 1em 0;
+    }
+    .tree-controls button {
+      margin-right: 6px;
+      padding: 6px 14px;
+      cursor: pointer;
+      background-color: rgba(138, 43, 226, 0.7);
+      color: white;
+      border: none;
+      border-radius: 4px;
+      transition: background-color 0.3s;
+    }
+    .tree-controls button:hover {
+      background-color: rgba(138, 43, 226, 0.9);
+    }
+  </style>
+</head>
+<body>
+  <!-- Header Overlay -->
+  <div class="header-overlay">
+    <div class="visualization-container">
+      <h1>Danbooru Tag Taxonomy</h1>
+      <h2 class="caption">Interactive hierarchical tree of Danbooru's tag classification system (~39,000 tags across 200 categories)</h2>
+
+      <div class="nav-buttons"></div>
+      <button class="nav-button" onclick="location.href='index.html'">Home</button>
+      <button class="nav-button" onclick="location.href='figure_15.html'">Next: Textual Training Data</button>
+
+      <div class="tree-controls">
+        <button id="expandAllBtn">Expand All Categories</button>
+        <button id="collapseAllBtn">Collapse All</button>
+        <button id="downloadBtn">Download SVG</button>
+      </div>
+    </div>
+  </div>
+
+  <svg class="tree-svg"></svg>
+
+  <script>
+    // Default to light mode
+    document.documentElement.classList.add('light-mode');
+
+    d3.json("json/danbooru_flat_full.json").then(function(data) {
+
+      // ---- CONFIG ----
+      const margin = { top: 20, right: 250, bottom: 20, left: 80 };
+      const nodeWidth = 220;
+      const duration = 400;
+
+      const svg = d3.select("svg.tree-svg");
+      const g = svg.append("g")
+        .attr("transform", `translate(${margin.left},${margin.top})`);
+
+      const linkGroup = g.append("g").attr("class", "links");
+      const nodeGroup = g.append("g").attr("class", "nodes");
+
+      // ---- STEP 0: Distinct color per top-level category ----
+      const categoryColors = {
+        "attire_and_body_accessories": "#e6194b",
+        "body":                        "#f58231",
+        "characters":                  "#3cb44b",
+        "copyrights_artists_projects_and_media": "#4363d8",
+        "creatures":                   "#42d4f4",
+        "drawing_software":            "#911eb4",
+        "games":                       "#f032e6",
+        "metatags":                    "#bfef45",
+        "more":                        "#9A6324",
+        "objects":                     "#e6beff",
+        "plant":                       "#006400",
+        "real_world":                  "#ffd8b1",
+        "sex":                         "#ff69b4",
+        "visual_characteristics":      "#000075",
+        "subject":                     "#ffe119",
+        "uncategorized":               "#a9a9a9",
+        "actions_and_expressions":     "#800000",
+        "objects_and_backgrounds":     "#469990",
+      };
+
+      function assignColors(node, inheritedColor) {
+        if (inheritedColor) node.color = inheritedColor;
+        if (node.children) node.children.forEach(child => assignColors(child, node.color));
+      }
+      if (data.children) {
+        data.children.forEach(child => {
+          child.color = categoryColors[child.name] || "#888";
+          assignColors(child, child.color);
+        });
+      }
+
+      // ---- Pre-process: sort + convert tags to leaf nodes ----
+      function preprocess(node) {
+        if (node.children && node.children.length > 0) {
+          node.children.forEach(preprocess);
+          node.children.sort((a, b) => (b.tag_count || 0) - (a.tag_count || 0));
+        }
+        else if (node.tags && node.tags.length > 0) {
+          node.children = node.tags
+            .filter(t => t !== "...")
+            .map(tag => ({
+              name: tag,
+              _isTag: true,
+              color: node.color,
+              tag_count: 0,
+              category_count: 0
+            }));
+          node.children.sort((a, b) => a.name.localeCompare(b.name));
+        }
+        return node;
+      }
+      preprocess(data);
+
+      function getColor(d) {
+        return d.data.color || "#888";
+      }
+
+      // ---- Build hierarchy ----
+      const root = d3.hierarchy(data, d => d.children);
+      root.x0 = 0;
+      root.y0 = 0;
+
+      let idCounter = 0;
+      root.each(d => { d.id = ++idCounter; });
+
+      // ---- Size scale for tag_count → radius ----
+      const allTagCounts = root.descendants()
+        .filter(d => d.depth > 0 && !d.data._isTag)
+        .map(d => d.data.tag_count || 0);
+      const sizeScale = d3.scaleSqrt()
+        .domain([0, d3.max(allTagCounts)])
+        .range([4, 18]);
+
+      function nodeRadius(d) {
+        if (d.data._isTag) return 2.5;
+        if (d.depth === 0) return 8;
+        return sizeScale(d.data.tag_count || 0);
+      }
+
+      function collapse(d) {
+        if (d.children) {
+          d._children = d.children;
+          d._children.forEach(collapse);
+          d.children = null;
+        }
+      }
+
+      function expandCategories(d) {
+        if (d._children) {
+          if (!d._children[0].data._isTag) {
+            d.children = d._children;
+            d._children = null;
+          }
+        }
+        if (d.children) d.children.forEach(expandCategories);
+      }
+
+      function expandAll(d) {
+        if (d._children) {
+          d.children = d._children;
+          d._children = null;
+        }
+        if (d.children) d.children.forEach(expandAll);
+      }
+
+      // Collapse everything below root initially
+      if (root.children) {
+        root.children.forEach(collapse);
+      }
+
+      function computeTreeLayout() {
+        const leaves = root.descendants().filter(d =>
+          !d.children || d.children.length === 0
+        );
+        const visibleCount = leaves.length;
+        const rowHeight = 18;
+        const treeHeight = Math.max(300, visibleCount * rowHeight);
+
+        const treeLayout = d3.tree()
+          .size([treeHeight, 0])
+          .separation((a, b) => {
+            if (a.data._isTag && b.data._isTag) return 0.6;
+            return a.parent === b.parent ? 1 : 1.3;
+          });
+
+        treeLayout(root);
+
+        root.each(d => { d.y = d.depth * nodeWidth; });
+
+        const maxDepth = d3.max(root.descendants(), d => d.depth) || 0;
+        const totalWidth = margin.left + margin.right + (maxDepth + 1) * nodeWidth;
+        const totalHeight = margin.top + margin.bottom + treeHeight;
+        svg.attr("width", totalWidth)
+           .attr("height", totalHeight)
+           .attr("viewBox", `0 0 ${totalWidth} ${totalHeight}`);
+      }
+
+      // ---- MAIN UPDATE FUNCTION ----
+      function update(source) {
+        computeTreeLayout();
+
+        const nodes = root.descendants();
+        const links = root.links();
+
+        // Links
+        const link = linkGroup.selectAll("path.link")
+          .data(links, d => d.target.id);
+
+        const linkEnter = link.enter()
+          .append("path")
+          .attr("class", "link")
+          .attr("d", () => {
+            const o = { x: source.x0, y: source.y0 };
+            return diagonal({ source: o, target: o });
+          });
+
+        linkEnter.merge(link).transition().duration(duration)
+          .attr("d", d => diagonal(d))
+          .attr("stroke", d => d.target.data.color || "#ccc")
+          .attr("stroke-opacity", 0.5);
+
+        link.exit().transition().duration(duration)
+          .attr("d", () => {
+            const o = { x: source.x, y: source.y };
+            return diagonal({ source: o, target: o });
+          })
+          .remove();
+
+        // Nodes
+        const node = nodeGroup.selectAll("g.node")
+          .data(nodes, d => d.id);
+
+        const nodeEnter = node.enter()
+          .append("g")
+          .attr("class", d => d.data._isTag ? "node leaf" : "node")
+          .attr("transform", () => `translate(${source.y0},${source.x0})`)
+          .on("click", (event, d) => {
+            if (d.data._isTag) return;
+            if (d.children) {
+              d._children = d.children;
+              d.children = null;
+            } else if (d._children) {
+              d.children = d._children;
+              d._children = null;
+            }
+            update(d);
+          });
+
+        nodeEnter.append("circle").attr("r", 1e-6);
+
+        nodeEnter.append("text")
+          .attr("dy", "0.32em")
+          .attr("x", d => nodeRadius(d) + 4)
+          .text(d => {
+            if (d.data._isTag) return d.data.name;
+            const count = d.data.tag_count || 0;
+            return count > 0 ? `${d.data.name} (${count})` : d.data.name;
+          });
+
+        nodeEnter.append("title");
+
+        const nodeUpdate = nodeEnter.merge(node);
+
+        nodeUpdate.transition().duration(duration)
+          .attr("transform", d => `translate(${d.y},${d.x})`);
+
+        nodeUpdate.select("circle")
+          .attr("r", d => nodeRadius(d))
+          .style("fill", d => {
+            if (d.data._isTag) return getColor(d);
+            return (d._children && d._children.length) ? getColor(d) : "#fff";
+          })
+          .style("stroke", d => getColor(d));
+
+        nodeUpdate.select("text")
+          .style("fill-opacity", 1)
+          .style("font-weight", d => {
+            if (d.data._isTag) return "normal";
+            return d.depth <= 1 ? "bold" : "normal";
+          })
+          .style("font-size", d => d.data._isTag ? "10px" : "12px")
+          .style("fill", d => d.data._isTag ? "#666" : "#333");
+
+        nodeUpdate.select("title")
+          .text(d => {
+            if (d.data._isTag) return d.data.name;
+            const parts = [d.data.name];
+            if (d.data.tag_count) parts.push(`Tags: ${d.data.tag_count}`);
+            if (d.data.category_count) parts.push(`Sub-categories: ${d.data.category_count}`);
+            const hidden = d._children ? d._children.length : 0;
+            if (hidden) {
+              const type = d._children[0].data._isTag ? "tags" : "children";
+              parts.push(`Click to expand (${hidden} ${type})`);
+            } else if (d.children && d.children.length) {
+              parts.push("Click to collapse");
+            }
+            return parts.join("\n");
+          });
+
+        const nodeExit = node.exit().transition().duration(duration)
+          .attr("transform", () => `translate(${source.y},${source.x})`)
+          .remove();
+
+        nodeExit.select("circle").attr("r", 1e-6);
+        nodeExit.select("text").style("fill-opacity", 1e-6);
+
+        nodes.forEach(d => { d.x0 = d.x; d.y0 = d.y; });
+      }
+
+      function diagonal(d) {
+        return `M${d.source.y},${d.source.x}
+                C${(d.source.y + d.target.y) / 2},${d.source.x}
+                 ${(d.source.y + d.target.y) / 2},${d.target.x}
+                 ${d.target.y},${d.target.x}`;
+      }
+
+      // Initial render
+      update(root);
+
+      // Expand All (categories only)
+      document.getElementById("expandAllBtn").addEventListener("click", () => {
+        expandCategories(root);
+        update(root);
+      });
+
+      // Collapse All
+      document.getElementById("collapseAllBtn").addEventListener("click", () => {
+        if (root.children) root.children.forEach(collapse);
+        update(root);
+      });
+
+      // Download SVG
+      document.getElementById("downloadBtn").addEventListener("click", () => {
+        const svgNode = document.querySelector("svg");
+        const clonedSvg = svgNode.cloneNode(true);
+        clonedSvg.setAttribute("xmlns", "http://www.w3.org/2000/svg");
+
+        const styleEl = document.createElement("style");
+        styleEl.textContent = `
+          .link { fill: none; stroke: #ccc; stroke-width: 1.5px; }
+          .node circle { stroke-width: 1.5px; }
+          .node text { font-size: 12px; fill: #333; font-family: sans-serif; }
+          .node.leaf text { font-size: 10px; fill: #666; }
+        `;
+        clonedSvg.insertBefore(styleEl, clonedSvg.firstChild);
+
+        const svgData = new XMLSerializer().serializeToString(clonedSvg);
+        const svgBlob = new Blob([svgData], { type: "image/svg+xml;charset=utf-8" });
+        const url = URL.createObjectURL(svgBlob);
+        const a = document.createElement("a");
+        a.href = url;
+        a.download = "danbooru_tree_interactive.svg";
+        document.body.appendChild(a);
+        a.click();
+        document.body.removeChild(a);
+        URL.revokeObjectURL(url);
+      });
+
+    });
+  </script>
+</body>
+</html>

public/Figure_8a_barchart.html

@@ -135,16 +135,30 @@ d3.json("json/sunburst_gender_A.json").then(data => {
     .attr("transform", `translate(${margin.left},${margin.top})`);
 
   /* --------------------------------------------------------
-     5. STACKED DATA
+     5. PREPARE DATA WITH SORTED SEGMENTS PER GENDER
      -------------------------------------------------------- */
-  const stack = d3.stack()
-    .keys(professionOrder)
-    .value((d, key) => {
-      const prof = d[1].professions.find(p => p.profession === key);
-      return prof ? prof.value : 0;
+  // Build stacked data manually with segments sorted by size per gender
+  const stackedData = [];
+  
+  genders.forEach(gender => {
+    const genderInfo = genderData.get(gender);
+    const professions = genderInfo.professions;
+    
+    // Sort professions by value descending (largest at bottom)
+    const sortedProfs = [...professions].sort((a, b) => b.value - a.value);
+    
+    let y0 = 0;
+    sortedProfs.forEach(prof => {
+      stackedData.push({
+        gender: gender,
+        profession: prof.profession,
+        value: prof.value,
+        y0: y0,
+        y1: y0 + prof.value
+      });
+      y0 += prof.value;
     });
-
-  const series = stack(Array.from(genderData));
+  });
 
   /* --------------------------------------------------------
      6. AXES
@@ -178,29 +192,22 @@ d3.json("json/sunburst_gender_A.json").then(data => {
     .text("Count");
 
   /* --------------------------------------------------------
-     7. DRAW BARS
+     7. DRAW BARS (using manually stacked data)
      -------------------------------------------------------- */
-  svg.append("g")
-    .selectAll("g")
-    .data(series)
-    .join("g")
-    .attr("fill", d => colorMap[d.key])
-    .selectAll("rect")
-    .data(d => d)
+  svg.selectAll(".bar")
+    .data(stackedData)
     .join("rect")
     .attr("class", "bar")
-    .attr("x", d => x(d.data[0]))
-    .attr("y", d => y(d[1]))
-    .attr("height", d => y(d[0]) - y(d[1]))
+    .attr("x", d => x(d.gender))
+    .attr("y", d => y(d.y1))
+    .attr("height", d => y(d.y0) - y(d.y1))
     .attr("width", x.bandwidth())
+    .attr("fill", d => colorMap[d.profession])
     .append("title")
-    .text(d => {
-      const profKey = series.find(s => s.includes(d))?.key;
-      return `${d.data[0]} – ${profKey}: ${d[1] - d[0]}`;
-    });
+    .text(d => `${d.gender} – ${d.profession}: ${d.value}`);
 
   /* --------------------------------------------------------
-     8. LEGEND (ordered by occurrence)
+     8. LEGEND (ordered by occurrence - unchanged)
      -------------------------------------------------------- */
   const legend = svg.append("g")
     .attr("transform", `translate(${width + 20}, 0)`);

public/Figure_8b_barchart.html

@@ -62,7 +62,7 @@ const colorMap = {
 };
 
 /* --------------------------------------------------------
-   2. LOAD JSON + TRANSFORM IT
+   3. LOAD JSON + TRANSFORM IT
    -------------------------------------------------------- */
 d3.json("json/sunburst_countries_A.json").then(data => {
 
@@ -81,25 +81,25 @@ d3.json("json/sunburst_countries_A.json").then(data => {
   });
 
   /* --------------------------------------------------------
-     3. CALCULATE PROFESSION ORDER (Other last)
+     3a. CALCULATE PROFESSION ORDER BY TOTAL OCCURRENCE
      -------------------------------------------------------- */
   const professionCounts = {};
   flatData.forEach(d => {
     professionCounts[d.profession] = (professionCounts[d.profession] || 0) + d.value;
   });
 
+  // Sort professions by count (descending), but always put "Other" last
   const professionOrder = Object.entries(professionCounts)
     .filter(entry => entry[0] !== "Other")
     .sort((a, b) => b[1] - a[1])
     .map(entry => entry[0]);
-
+  
+  // Add "Other" at the end if it exists
   if (professionCounts["Other"]) {
     professionOrder.push("Other");
   }
 
-  /* --------------------------------------------------------
-     4. GROUP DATA BY COUNTRY
-     -------------------------------------------------------- */
+  // Group per country
   const countryData = d3.rollup(
     flatData,
     v => ({
@@ -110,22 +110,14 @@ d3.json("json/sunburst_countries_A.json").then(data => {
   );
 
   /* --------------------------------------------------------
-     5. SORT COUNTRIES BY TOTAL — but put "Other" last
+     NEW: Sort countries by total occurrences (descending)
      -------------------------------------------------------- */
-  let countries = Array.from(countryData.entries())
+  const countries = Array.from(countryData.entries())
     .sort((a, b) => b[1].total - a[1].total)
     .map(d => d[0]);
 
-  // Move country named "Other" to the final position
-  countries = countries.filter(c => c !== "Other");
-  if (countryData.has("Other")) {
-    countries.push("Other");
-  }
-
-  const topCountries = countries.slice(0, 25);
-
   /* --------------------------------------------------------
-     6. SVG SETUP
+     4. SVG SETUP
      -------------------------------------------------------- */
   const margin = {top: 40, right: 200, bottom: 150, left: 80};
   const width = Math.max(800, window.innerWidth - 100) - margin.left - margin.right;
@@ -138,19 +130,35 @@ d3.json("json/sunburst_countries_A.json").then(data => {
     .attr("transform", `translate(${margin.left},${margin.top})`);
 
   /* --------------------------------------------------------
-     7. STACK LAYOUT
+     5. PREPARE DATA WITH SORTED SEGMENTS PER COUNTRY
      -------------------------------------------------------- */
-  const stack = d3.stack()
-    .keys(professionOrder)
-    .value((d, key) => {
-      const prof = d[1].professions.find(p => p.profession === key);
-      return prof ? prof.value : 0;
+  const topCountries = countries.slice(0, 25);
+  
+  // Build stacked data manually with segments sorted by size per country
+  const stackedData = [];
+  
+  topCountries.forEach(country => {
+    const countryInfo = countryData.get(country);
+    const professions = countryInfo.professions;
+    
+    // Sort professions by value descending (largest at bottom)
+    const sortedProfs = [...professions].sort((a, b) => b.value - a.value);
+    
+    let y0 = 0;
+    sortedProfs.forEach(prof => {
+      stackedData.push({
+        country: country,
+        profession: prof.profession,
+        value: prof.value,
+        y0: y0,
+        y1: y0 + prof.value
+      });
+      y0 += prof.value;
     });
-
-  const series = stack(Array.from(countryData));
+  });
 
   /* --------------------------------------------------------
-     8. AXES
+     6. AXES
      -------------------------------------------------------- */
   const x = d3.scaleBand()
     .domain(topCountries)
@@ -183,29 +191,22 @@ d3.json("json/sunburst_countries_A.json").then(data => {
     .text("Count");
 
   /* --------------------------------------------------------
-     9. DRAW STACKED BARS
+     7. DRAW BARS (using manually stacked data)
      -------------------------------------------------------- */
-  svg.append("g")
-    .selectAll("g")
-    .data(series)
-    .join("g")
-    .attr("fill", d => colorMap[d.key])
-    .selectAll("rect")
-    .data(d => d)
+  svg.selectAll(".bar")
+    .data(stackedData)
     .join("rect")
     .attr("class", "bar")
-    .attr("x", d => x(d.data[0]))
-    .attr("y", d => y(d[1]))
-    .attr("height", d => y(d[0]) - y(d[1]))
+    .attr("x", d => x(d.country))
+    .attr("y", d => y(d.y1))
+    .attr("height", d => y(d.y0) - y(d.y1))
     .attr("width", x.bandwidth())
+    .attr("fill", d => colorMap[d.profession])
     .append("title")
-    .text(d => {
-      const profKey = series.find(s => s.includes(d))?.key;
-      return `${d.data[0]} – ${profKey}: ${d[1] - d[0]}`;
-    });
+    .text(d => `${d.country} – ${d.profession}: ${d.value}`);
 
   /* --------------------------------------------------------
-     10. LEGEND
+     8. LEGEND (unchanged - uses original professionOrder)
      -------------------------------------------------------- */
   const legend = svg.append("g")
     .attr("transform", `translate(${width + 20}, 0)`);
@@ -215,18 +216,18 @@ d3.json("json/sunburst_countries_A.json").then(data => {
 
     row.append("rect")
       .attr("width", 18)
-      .attr("height": 18)
-      .attr("fill": colorMap[prof]);
+      .attr("height", 18)
+      .attr("fill", colorMap[prof]);
 
     row.append("text")
-      .attr("x": 24)
-      .attr("y": 9)
-      .attr("dy": "0.35em")
+      .attr("x", 24)
+      .attr("y", 9)
+      .attr("dy", "0.35em")
       .text(`${prof} (${professionCounts[prof] || 0})`);
   });
 
   /* --------------------------------------------------------
-     11. DOWNLOAD SVG BUTTON
+     9. DOWNLOAD SVG BUTTON
      -------------------------------------------------------- */
   document.getElementById("downloadBtn").addEventListener("click", () => {
     const svgNode = document.querySelector("#chart");
@@ -254,6 +255,5 @@ d3.json("json/sunburst_countries_A.json").then(data => {
 });
 </script>
 
-
 </body>
 </html>

public/figure_15.html

@@ -1,5 +1,5 @@
 <!DOCTYPE html>
-<html lang="en">
+<html lang="en" class="light-mode">
 <head>
   <meta charset="utf-8">
   <script src="https://d3js.org/d3.v7.min.js"></script>

public/figure_5.html

@@ -1,5 +1,5 @@
 <!DOCTYPE html>
-<html lang="en">
+<html lang="en" class="light-mode">
 <head>
   <meta charset="utf-8">
   <script src="https://d3js.org/d3.v7.min.js"></script>
@@ -401,7 +401,7 @@
     const toggle = document.getElementById('modeToggle');
     const root = document.documentElement;
 
-    root.classList.remove('light-mode');
+    root.classList.add('light-mode');
 
     toggle.addEventListener('change', () => {
       root.classList.toggle('light-mode');

public/index.html

@@ -22,11 +22,6 @@
     </p>
   </div>
   
-  <div class="theme-toggle" style="text-align: center; margin-bottom: 2rem;">
-    <label>
-      <input type="checkbox" id="modeToggle"> Toggle Light Mode
-    </label>
-  </div>
   <div class="card-grid">
     <a class="card-link" href="figure_15.html">
       <div class="preview-card">
@@ -35,7 +30,7 @@
         <div class="card-footer">View Graph →</div>
       </div>
     </a>
-    
+
     <a class="card-link" href="figure_5.html">
       <div class="preview-card">
         <img src="img/japan.svg" alt="Preview 3">
@@ -43,19 +38,20 @@
         <div class="card-footer">Explore Tags →</div>
       </div>
     </a>
+
+    <a class="card-link" href="Figure_13_interactive.html">
+      <div class="preview-card">
+        <img src="img/danbooru_tree_interactive.svg" alt="Preview 3">
+        <!-- <div style="width:100%;height:100%;display:flex;align-items:center;justify-content:center;background:#f0f0f0;color:#888;font-size:3rem;">🌳</div> -->
+        <div class="card-title">Danbooru Tag Taxonomy</div>
+        <div class="card-footer">Explore Tree →</div>
+      </div>
+    </a>
   </div>
-  
-</div>
+
   <script>
-    const toggle = document.getElementById('modeToggle');
-    const root = document.documentElement;
-  
-    // Default to dark mode
-    root.classList.remove('light-mode');
-  
-    toggle.addEventListener('change', () => {
-      root.classList.toggle('light-mode');
-    });
+    // Default to light mode
+    document.documentElement.classList.add('light-mode');
   </script>
   
 </body>

public/styles.css

@@ -159,7 +159,7 @@ input[type=range] {
   display: grid;
   grid-template-columns: repeat(auto-fit, minmax(220px, 1fr));
   gap: 2rem;
-  max-width: 700px;
+  max-width: 900px;
   margin: 2rem auto;
   padding: 0 1rem;
   justify-content: center;