{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5cfa8ac4",
   "metadata": {},
   "source": [
    "# 03 — Typology Modeling\n",
    "\n",
    "**Reproduksi StandardScaler → PCA → K-Means dan audit robustness**\n",
    "\n",
    "Notebook ini adalah POC portfolio yang dapat dijalankan ulang dari artefak lokal Pasuruan Lens. Kode produksi tetap berada di `python/scripts/` dan `python/src/pasuruan365/`; notebook berfungsi sebagai narasi analitik yang ringkas, transparan, dan mudah dipresentasikan."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "87393d49",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:20.756537Z",
     "iopub.status.busy": "2026-09-04T06:27:20.756379Z",
     "iopub.status.idle": "2026-09-04T06:27:22.222290Z",
     "shell.execute_reply": "2026-09-04T06:27:22.221418Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Project root: PASURUAN365\n",
      "Python: 3.12.6\n"
     ]
    }
   ],
   "source": [
    "from pathlib import Path\n",
    "import json\n",
    "import sys\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "\n",
    "def find_project_root(start: Path | None = None) -> Path:\n",
    "    start = (start or Path.cwd()).resolve()\n",
    "    for candidate in (start, *start.parents):\n",
    "        if (candidate / \"package.json\").exists() and (candidate / \"data\").exists():\n",
    "            return candidate\n",
    "    raise RuntimeError(\"Root Pasuruan Lens tidak ditemukan. Jalankan notebook dari repository ini.\")\n",
    "\n",
    "\n",
    "ROOT = find_project_root()\n",
    "sys.path.insert(0, str(ROOT / \"python\" / \"src\"))\n",
    "\n",
    "pd.set_option(\"display.max_columns\", 100)\n",
    "pd.set_option(\"display.max_colwidth\", 100)\n",
    "sns.set_theme(style=\"whitegrid\", context=\"notebook\")\n",
    "COLORS = {\"A\": \"#2563eb\", \"B\": \"#f59e0b\", \"C\": \"#10b981\"}\n",
    "\n",
    "print(f\"Project root: {ROOT}\")\n",
    "print(f\"Python: {sys.version.split()[0]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f80e072a",
   "metadata": {},
   "source": [
    "## Pertanyaan modeling\n",
    "\n",
    "Bisakah 365 desa/kelurahan dikelompokkan berdasarkan kemiripan struktural lintas domain? Hasilnya adalah **tipologi eksploratif**, bukan ranking pembangunan dan bukan label benar/salah.\n",
    "\n",
    "Model V2 memilih delapan fitur dari enam domain, StandardScaler, PCA tujuh komponen, dan K-Means `k=3` setelah membandingkan beberapa spesifikasi serta sensitivitas model V1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "65497698",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:22.225152Z",
     "iopub.status.busy": "2026-09-04T06:27:22.224636Z",
     "iopub.status.idle": "2026-09-04T06:27:23.254367Z",
     "shell.execute_reply": "2026-09-04T06:27:23.253879Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>feature_name</th>\n",
       "      <th>domain</th>\n",
       "      <th>model_role</th>\n",
       "      <th>transform</th>\n",
       "      <th>coverage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>population_share_of_district</td>\n",
       "      <td>DEMOGRAPHY</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sex_ratio</td>\n",
       "      <td>DEMOGRAPHY</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>log1p_population_density</td>\n",
       "      <td>GEOGRAPHY</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>ALREADY_LOG1P</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>log1p_bts_per_10000_population</td>\n",
       "      <td>CONNECTIVITY</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>ALREADY_LOG1P</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>cellular_operator_count</td>\n",
       "      <td>CONNECTIVITY</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>has_banking_facility</td>\n",
       "      <td>ECONOMY</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>any_documented_disaster_event_2024</td>\n",
       "      <td>DISASTER</td>\n",
       "      <td>ROBUST_CORE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>has_polyclinic</td>\n",
       "      <td>HEALTHCARE</td>\n",
       "      <td>BINARY_PRESENCE_CORE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>365/365</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         feature_name        domain            model_role  \\\n",
       "0        population_share_of_district    DEMOGRAPHY           ROBUST_CORE   \n",
       "1                           sex_ratio    DEMOGRAPHY           ROBUST_CORE   \n",
       "2            log1p_population_density     GEOGRAPHY           ROBUST_CORE   \n",
       "3      log1p_bts_per_10000_population  CONNECTIVITY           ROBUST_CORE   \n",
       "4             cellular_operator_count  CONNECTIVITY           ROBUST_CORE   \n",
       "5                has_banking_facility       ECONOMY           ROBUST_CORE   \n",
       "6  any_documented_disaster_event_2024      DISASTER           ROBUST_CORE   \n",
       "7                      has_polyclinic    HEALTHCARE  BINARY_PRESENCE_CORE   \n",
       "\n",
       "       transform coverage  \n",
       "0           NONE  365/365  \n",
       "1           NONE  365/365  \n",
       "2  ALREADY_LOG1P  365/365  \n",
       "3  ALREADY_LOG1P  365/365  \n",
       "4           NONE  365/365  \n",
       "5           NONE  365/365  \n",
       "6           NONE  365/365  \n",
       "7           NONE  365/365  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.metrics import adjusted_rand_score, silhouette_score\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "eda = pd.read_csv(ROOT / \"data/analytics/village_eda_features.csv\", dtype={\"village_id\": str})\n",
    "saved = pd.read_csv(ROOT / \"data/modeling/cluster_assignments_v2.csv\", dtype={\"village_id\": str})\n",
    "config = json.loads((ROOT / \"data/modeling/selected_model_config_v2.json\").read_text())\n",
    "manifest = pd.read_csv(ROOT / \"data/modeling/review/model_feature_manifest_v2.csv\")\n",
    "\n",
    "manifest[[\"feature_name\", \"domain\", \"model_role\", \"transform\", \"coverage\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e2c7eb1",
   "metadata": {},
   "source": [
    "## POC 1 — Reproduksi pipeline model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c20dd232",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:23.256410Z",
     "iopub.status.busy": "2026-09-04T06:27:23.256195Z",
     "iopub.status.idle": "2026-09-04T06:27:23.411277Z",
     "shell.execute_reply": "2026-09-04T06:27:23.410409Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nilai</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Adjusted Rand Index vs frozen assignment</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Silhouette reproduced</th>\n",
       "      <td>0.175881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Silhouette frozen config</th>\n",
       "      <td>0.175881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PCA variance retained</th>\n",
       "      <td>0.938386</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                             nilai\n",
       "Adjusted Rand Index vs frozen assignment  1.000000\n",
       "Silhouette reproduced                     0.175881\n",
       "Silhouette frozen config                  0.175881\n",
       "PCA variance retained                     0.938386"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_frame = eda.copy()\n",
    "model_frame[\"has_polyclinic\"] = model_frame[\"polyclinic_count\"].gt(0).astype(int)\n",
    "X = model_frame[config[\"features\"]].astype(float)\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X)\n",
    "pca = PCA(n_components=config[\"pca_components\"], random_state=config[\"random_state\"])\n",
    "X_pca = pca.fit_transform(X_scaled)\n",
    "model = KMeans(n_clusters=config[\"k\"], random_state=config[\"random_state\"], n_init=20)\n",
    "reproduced_labels = model.fit_predict(X_pca)\n",
    "\n",
    "comparison = model_frame[[\"village_id\"]].merge(saved[[\"village_id\", \"cluster_id\"]], on=\"village_id\")\n",
    "ari = adjusted_rand_score(comparison[\"cluster_id\"], reproduced_labels)\n",
    "pd.Series({\n",
    "    \"Adjusted Rand Index vs frozen assignment\": ari,\n",
    "    \"Silhouette reproduced\": silhouette_score(X_pca, reproduced_labels),\n",
    "    \"Silhouette frozen config\": config[\"metrics\"][\"silhouette\"],\n",
    "    \"PCA variance retained\": pca.explained_variance_ratio_.sum(),\n",
    "}).round(6).to_frame(\"nilai\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e3aa959e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:23.414383Z",
     "iopub.status.busy": "2026-09-04T06:27:23.413921Z",
     "iopub.status.idle": "2026-09-04T06:27:23.418591Z",
     "shell.execute_reply": "2026-09-04T06:27:23.417684Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "✓ Assignment V2 berhasil direproduksi (ARI permutation-invariant ≈ 1).\n"
     ]
    }
   ],
   "source": [
    "assert ari > 0.999, \"Reproduksi model berbeda dari artefak V2.\"\n",
    "print(\"✓ Assignment V2 berhasil direproduksi (ARI permutation-invariant ≈ 1).\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1ae58a9",
   "metadata": {},
   "source": [
    "## POC 2 — PCA adalah representasi model, PC1–PC2 hanya visualisasi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "72823afd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:23.421587Z",
     "iopub.status.busy": "2026-09-04T06:27:23.421333Z",
     "iopub.status.idle": "2026-09-04T06:27:23.788522Z",
     "shell.execute_reply": "2026-09-04T06:27:23.787751Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>component</th>\n",
       "      <th>explained</th>\n",
       "      <th>cumulative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.275</td>\n",
       "      <td>0.275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0.159</td>\n",
       "      <td>0.434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0.129</td>\n",
       "      <td>0.562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0.111</td>\n",
       "      <td>0.673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0.109</td>\n",
       "      <td>0.782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0.088</td>\n",
       "      <td>0.870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.938</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   component  explained  cumulative\n",
       "0          1      0.275       0.275\n",
       "1          2      0.159       0.434\n",
       "2          3      0.129       0.562\n",
       "3          4      0.111       0.673\n",
       "4          5      0.109       0.782\n",
       "5          6      0.088       0.870\n",
       "6          7      0.069       0.938"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "variance = pd.DataFrame({\n",
    "    \"component\": np.arange(1, len(pca.explained_variance_ratio_) + 1),\n",
    "    \"explained\": pca.explained_variance_ratio_,\n",
    "})\n",
    "variance[\"cumulative\"] = variance[\"explained\"].cumsum()\n",
    "\n",
    "ax = variance.plot(x=\"component\", y=\"cumulative\", marker=\"o\", ylim=(0, 1.05), figsize=(8, 4), legend=False)\n",
    "ax.axhline(.80, color=\"#dc2626\", linestyle=\"--\", linewidth=1)\n",
    "ax.set(title=\"Cumulative explained variance\", xlabel=\"Jumlah komponen\", ylabel=\"Proporsi kumulatif\")\n",
    "plt.tight_layout()\n",
    "variance.round(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "611078e7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:23.790926Z",
     "iopub.status.busy": "2026-09-04T06:27:23.790720Z",
     "iopub.status.idle": "2026-09-04T06:27:24.066007Z",
     "shell.execute_reply": "2026-09-04T06:27:24.065468Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_frame = pd.DataFrame({\"PC1\": X_pca[:, 0], \"PC2\": X_pca[:, 1]})\n",
    "plot_frame[\"typology\"] = saved[\"cluster_label\"].str.replace(\"Typology \", \"\", regex=False)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8, 6))\n",
    "for label, group in plot_frame.groupby(\"typology\"):\n",
    "    ax.scatter(group.PC1, group.PC2, s=28, alpha=.72, label=f\"Tipologi {label}\", color=COLORS[label])\n",
    "ax.set(title=\"Proyeksi PC1–PC2 (bukan keseluruhan ruang model)\", xlabel=\"PC1\", ylabel=\"PC2\")\n",
    "ax.legend()\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19535fe0",
   "metadata": {},
   "source": [
    "## POC 3 — Separation harus dibaca bersama stability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "389b385c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:24.068529Z",
     "iopub.status.busy": "2026-09-04T06:27:24.068290Z",
     "iopub.status.idle": "2026-09-04T06:27:24.075104Z",
     "shell.execute_reply": "2026-09-04T06:27:24.074666Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nilai</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>silhouette</th>\n",
       "      <td>0.176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>seed_stability_ari</th>\n",
       "      <td>0.983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bootstrap_stability_ari</th>\n",
       "      <td>0.839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scaler_robustness</th>\n",
       "      <td>0.549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_ablation_robustness</th>\n",
       "      <td>0.622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cross_specification_robustness</th>\n",
       "      <td>0.615</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                nilai\n",
       "silhouette                      0.176\n",
       "seed_stability_ari              0.983\n",
       "bootstrap_stability_ari         0.839\n",
       "scaler_robustness               0.549\n",
       "feature_ablation_robustness     0.622\n",
       "cross_specification_robustness  0.615"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics = pd.Series(config[\"metrics\"])\n",
    "metrics[[\n",
    "    \"silhouette\", \"seed_stability_ari\", \"bootstrap_stability_ari\",\n",
    "    \"scaler_robustness\", \"feature_ablation_robustness\", \"cross_specification_robustness\",\n",
    "]].round(3).to_frame(\"nilai\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3d8c2144",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-04T06:27:24.077127Z",
     "iopub.status.busy": "2026-09-04T06:27:24.076918Z",
     "iopub.status.idle": "2026-09-04T06:27:24.084041Z",
     "shell.execute_reply": "2026-09-04T06:27:24.083527Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>desa_kelurahan</th>\n",
       "      <th>persen</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>kelas</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>HIGH_CONFIDENCE</th>\n",
       "      <td>214</td>\n",
       "      <td>58.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODERATE_AMBIGUITY</th>\n",
       "      <td>97</td>\n",
       "      <td>26.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HIGH_AMBIGUITY</th>\n",
       "      <td>54</td>\n",
       "      <td>14.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    desa_kelurahan  persen\n",
       "kelas                                     \n",
       "HIGH_CONFIDENCE                214    58.6\n",
       "MODERATE_AMBIGUITY              97    26.6\n",
       "HIGH_AMBIGUITY                  54    14.8"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ambiguity = saved[\"ambiguity_class\"].value_counts().rename_axis(\"kelas\").to_frame(\"desa_kelurahan\")\n",
    "ambiguity[\"persen\"] = (ambiguity[\"desa_kelurahan\"] / len(saved) * 100).round(1)\n",
    "ambiguity"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a0067ad",
   "metadata": {},
   "source": [
    "## Kesimpulan\n",
    "\n",
    "Model dapat direproduksi, seed/bootstrap stability kuat, tetapi robustness lintas scaler dan spesifikasi tidak sempurna. Karena itu UI mempertahankan kelas ambiguitas dan tidak mempresentasikan tipologi sebagai kebenaran absolut. Kesederhanaan `k=3` dipilih sebagai keputusan komunikasi sekaligus robustness, bukan sekadar mengejar silhouette tertinggi."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.6"
  },
  "pasuruan_lens": {
   "slug": "typology-modeling",
   "type": "portfolio_poc"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}