add stuff

README.md

@@ -12,7 +12,8 @@ CIVITAI_VISUALIZATIONS/
 │   ├── subset1/
 │   ├── subset2/
 │   └── ...
-├── misc/                                   # Miscellaneous scripts and utilities
+├── misc/
+│   ├── credentials/                        # store api keys (CivitAI, Deepseek) etc. here
 ├── plots/                                  # Generated plots and visualizations
 ├── scripts/                                # Jupyter notebooks for analysis
 │   ├── 0_Scraping_image_metadata.ipynb     # Scraping scripts interacting with the CivitAI API, dataset consolidation

public/json/sunburst_countries.json

@@ -0,0 +1,468 @@
+{
+  "name": "root",
+  "children": [
+    {
+      "name": "Australia",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 64
+        },
+        {
+          "name": "Adult Performer",
+          "value": 3
+        },
+        {
+          "name": "Model",
+          "value": 27
+        },
+        {
+          "name": "Online Personality",
+          "value": 3
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 7
+        },
+        {
+          "name": "Other",
+          "value": 11
+        }
+      ]
+    },
+    {
+      "name": "Brazil",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 38
+        },
+        {
+          "name": "Adult Performer",
+          "value": 10
+        },
+        {
+          "name": "Model",
+          "value": 29
+        },
+        {
+          "name": "Online Personality",
+          "value": 12
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 9
+        },
+        {
+          "name": "Other",
+          "value": 14
+        }
+      ]
+    },
+    {
+      "name": "Canada",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 96
+        },
+        {
+          "name": "Adult Performer",
+          "value": 19
+        },
+        {
+          "name": "Model",
+          "value": 15
+        },
+        {
+          "name": "Online Personality",
+          "value": 15
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 20
+        },
+        {
+          "name": "Voice Actor",
+          "value": 2
+        },
+        {
+          "name": "Other",
+          "value": 12
+        }
+      ]
+    },
+    {
+      "name": "China",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 145
+        },
+        {
+          "name": "Adult Performer",
+          "value": 2
+        },
+        {
+          "name": "Model",
+          "value": 10
+        },
+        {
+          "name": "Online Personality",
+          "value": 8
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 27
+        },
+        {
+          "name": "Voice Actor",
+          "value": 2
+        },
+        {
+          "name": "Other",
+          "value": 19
+        }
+      ]
+    },
+    {
+      "name": "France",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 41
+        },
+        {
+          "name": "Adult Performer",
+          "value": 12
+        },
+        {
+          "name": "Model",
+          "value": 11
+        },
+        {
+          "name": "Online Personality",
+          "value": 5
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 13
+        },
+        {
+          "name": "Other",
+          "value": 26
+        }
+      ]
+    },
+    {
+      "name": "Germany",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 26
+        },
+        {
+          "name": "Adult Performer",
+          "value": 16
+        },
+        {
+          "name": "Model",
+          "value": 27
+        },
+        {
+          "name": "Online Personality",
+          "value": 16
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 11
+        },
+        {
+          "name": "Other",
+          "value": 32
+        }
+      ]
+    },
+    {
+      "name": "India",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 127
+        },
+        {
+          "name": "Adult Performer",
+          "value": 2
+        },
+        {
+          "name": "Model",
+          "value": 8
+        },
+        {
+          "name": "Online Personality",
+          "value": 3
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 3
+        },
+        {
+          "name": "Other",
+          "value": 3
+        }
+      ]
+    },
+    {
+      "name": "Indonesia",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 15
+        },
+        {
+          "name": "Adult Performer",
+          "value": 5
+        },
+        {
+          "name": "Model",
+          "value": 13
+        },
+        {
+          "name": "Online Personality",
+          "value": 19
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 25
+        },
+        {
+          "name": "Other",
+          "value": 3
+        }
+      ]
+    },
+    {
+      "name": "Japan",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 145
+        },
+        {
+          "name": "Adult Performer",
+          "value": 371
+        },
+        {
+          "name": "Model",
+          "value": 52
+        },
+        {
+          "name": "Online Personality",
+          "value": 6
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 482
+        },
+        {
+          "name": "Voice Actor",
+          "value": 297
+        },
+        {
+          "name": "Other",
+          "value": 67
+        }
+      ]
+    },
+    {
+      "name": "Other",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 380
+        },
+        {
+          "name": "Adult Performer",
+          "value": 184
+        },
+        {
+          "name": "Model",
+          "value": 308
+        },
+        {
+          "name": "Online Personality",
+          "value": 145
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 151
+        },
+        {
+          "name": "Voice Actor",
+          "value": 6
+        },
+        {
+          "name": "Other",
+          "value": 318
+        }
+      ]
+    },
+    {
+      "name": "Russia",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 47
+        },
+        {
+          "name": "Adult Performer",
+          "value": 89
+        },
+        {
+          "name": "Model",
+          "value": 73
+        },
+        {
+          "name": "Online Personality",
+          "value": 22
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 24
+        },
+        {
+          "name": "Voice Actor",
+          "value": 3
+        },
+        {
+          "name": "Other",
+          "value": 41
+        }
+      ]
+    },
+    {
+      "name": "South Korea",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 85
+        },
+        {
+          "name": "Adult Performer",
+          "value": 1
+        },
+        {
+          "name": "Model",
+          "value": 14
+        },
+        {
+          "name": "Online Personality",
+          "value": 3
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 405
+        },
+        {
+          "name": "Other",
+          "value": 9
+        }
+      ]
+    },
+    {
+      "name": "Ukraine",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 13
+        },
+        {
+          "name": "Adult Performer",
+          "value": 26
+        },
+        {
+          "name": "Model",
+          "value": 35
+        },
+        {
+          "name": "Online Personality",
+          "value": 6
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 8
+        },
+        {
+          "name": "Other",
+          "value": 9
+        }
+      ]
+    },
+    {
+      "name": "United Kingdom",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 253
+        },
+        {
+          "name": "Adult Performer",
+          "value": 13
+        },
+        {
+          "name": "Model",
+          "value": 31
+        },
+        {
+          "name": "Online Personality",
+          "value": 16
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 45
+        },
+        {
+          "name": "Voice Actor",
+          "value": 2
+        },
+        {
+          "name": "Other",
+          "value": 65
+        }
+      ]
+    },
+    {
+      "name": "United States",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1041
+        },
+        {
+          "name": "Adult Performer",
+          "value": 894
+        },
+        {
+          "name": "Model",
+          "value": 204
+        },
+        {
+          "name": "Online Personality",
+          "value": 219
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 168
+        },
+        {
+          "name": "Voice Actor",
+          "value": 36
+        },
+        {
+          "name": "Other",
+          "value": 289
+        }
+      ]
+    }
+  ]
+}

public/json/sunburst_countries_A.json

@@ -0,0 +1,129 @@
+{
+  "name": "root",
+  "children": [
+    {
+      "name": "Australia",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "China",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Colombia",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Czechia",
+      "children": [
+        {
+          "name": "Voice Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Denmark",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Germany",
+      "children": [
+        {
+          "name": "Online Personality",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "India",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Italy",
+      "children": [
+        {
+          "name": "Other",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Japan",
+      "children": [
+        {
+          "name": "Voice Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "Other",
+      "children": [
+        {
+          "name": "Singer, Musician",
+          "value": 1
+        },
+        {
+          "name": "Other",
+          "value": 2
+        }
+      ]
+    },
+    {
+      "name": "Poland",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 1
+        }
+      ]
+    },
+    {
+      "name": "US",
+      "children": [
+        {
+          "name": "Actor",
+          "value": 2
+        },
+        {
+          "name": "Adult Performer",
+          "value": 2
+        },
+        {
+          "name": "Online Personality",
+          "value": 3
+        },
+        {
+          "name": "Singer, Musician",
+          "value": 1
+        }
+      ]
+    }
+  ]
+}

scripts/01_prepare_sankey_csv.ipynb

@@ -9,13 +9,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
     "import pandas as pd\n",
     "from pathlib import Path\n",
-    "\n",
     "current_dir = Path.cwd()"
    ]
   },
@@ -45,6 +44,8 @@
     }
    ],
    "source": [
+    "import pandas as pd\n",
+    "# Load your dataset\n",
     "df = pd.read_csv(\"data/CSV/all_models_with_tags.csv\", nrows=40000)\n",
     "\n",
     "\n",

scripts/{05_countries_professions_victims.ipynb → Section_3-3-4_Figure_8_deepfake_victims.ipynb}

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -17,12 +17,16 @@
     "import pandas as pd\n",
     "from collections import defaultdict\n",
     "import json\n",
+    "from pathlib import Path\n",
     "\n",
-    "# Load the dataset\n",
-    "df = pd.read_csv(\"/home/lauhp/000_PHD/000_000_RESEARCH/000_X_Research_phase_1/d3js/data/model_subsets/POI/countries_professions.csv\")\n",
+    "current_dir = Path.cwd()\n",
     "\n",
-    "# ---- Step 1: Limit to top 10 countries ----\n",
-    "# Force 'Unknown' and empty countries to 'Other' before computing top countries\n",
+    "sunburst_path = current_dir.parent / \"public/json/sunburst_countries_A.json\"\n",
+    "\n",
+    "\n",
+    "aggregated_poi = current_dir.parent / \"data/CSV/Deepseek_annotated_POI_aggregated.csv\"\n",
+    "\n",
+    "df = pd.read_csv(aggregated_poi)\n",
     "df['country_cleaned'] = df['country'].apply(lambda x: x if x not in ['Unknown', '', None] else 'Other')\n",
     "\n",
     "# Now get top countries excluding what was forced into 'Other'\n",
@@ -68,9 +72,10 @@
     "    sunburst_dict[\"children\"].append({\"name\": country, \"children\": professions})\n",
     "\n",
     "# ---- Step 5: Save to a JSON file ----\n",
-    "with open(\"public/json/sunburst_data.json\", \"w\", encoding='utf-8') as f:\n",
+    "with open(sunburst_path, \"w\", encoding='utf-8') as f:\n",
     "    json.dump(sunburst_dict, f, ensure_ascii=False, indent=2)\n",
     "\n",
+    "\n",
     "print(\"✅ Sunburst data saved to sunburst_data.json\")\n"
    ]
   },