{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "rOpB1TRsPVhL",
        "outputId": "6a1a7201-cfae-40f6-a005-19bf7cbaa119"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "================================================================================\n",
            "QUANTITATIVE CYBERSECURITY RISK ASSESSMENT MODEL\n",
            "FOR HETEROGENEOUS IOT ENVIRONMENTS\n",
            "Implementation of Chapter 3 Methodology\n",
            "================================================================================\n",
            "\n",
            "================================================================================\n",
            "PHASE 1: DATA UNDERSTANDING\n",
            "================================================================================\n",
            "Loading CICIoT2023 dataset...\n",
            "Note: Using synthetic data for demonstration. Replace with actual CICIoT2023 dataset.\n",
            "\n",
            "============================================================\n",
            "DATA UNDERSTANDING PHASE\n",
            "============================================================\n",
            "\n",
            "Dataset Shape: (10000, 14)\n",
            "\n",
            "Columns: ['flow_duration', 'protocol', 'packet_rate', 'packet_count', 'bytes_sent', 'bytes_received', 'tcp_flags', 'packet_size_mean', 'packet_size_std', 'inter_arrival_time', 'flow_rate', 'source_port', 'destination_port', 'attack_category']\n",
            "\n",
            "Data Types:\n",
            "flow_duration          float64\n",
            "protocol                 int64\n",
            "packet_rate            float64\n",
            "packet_count             int64\n",
            "bytes_sent             float64\n",
            "bytes_received         float64\n",
            "tcp_flags                int64\n",
            "packet_size_mean       float64\n",
            "packet_size_std        float64\n",
            "inter_arrival_time     float64\n",
            "flow_rate              float64\n",
            "source_port              int64\n",
            "destination_port         int64\n",
            "attack_category       category\n",
            "dtype: object\n",
            "\n",
            "Missing Values:\n",
            "flow_duration         0\n",
            "protocol              0\n",
            "packet_rate           0\n",
            "packet_count          0\n",
            "bytes_sent            0\n",
            "bytes_received        0\n",
            "tcp_flags             0\n",
            "packet_size_mean      0\n",
            "packet_size_std       0\n",
            "inter_arrival_time    0\n",
            "flow_rate             0\n",
            "source_port           0\n",
            "destination_port      0\n",
            "attack_category       0\n",
            "dtype: int64\n",
            "\n",
            "Attack Category Distribution:\n",
            "attack_category\n",
            "Benign            7228\n",
            "Malicious         1318\n",
            "DoS                493\n",
            "Brute Force        312\n",
            "Spoofing           296\n",
            "Mirai              200\n",
            "DDoS               142\n",
            "Reconnaissance      11\n",
            "Name: count, dtype: int64\n",
            "\n",
            "Percentage Distribution:\n",
            "attack_category\n",
            "Benign            72.28\n",
            "Malicious         13.18\n",
            "DoS                4.93\n",
            "Brute Force        3.12\n",
            "Spoofing           2.96\n",
            "Mirai              2.00\n",
            "DDoS               1.42\n",
            "Reconnaissance     0.11\n",
            "Name: proportion, dtype: float64\n",
            "\n",
            "Numerical Features Summary:\n",
            "       flow_duration      protocol   packet_rate  packet_count    bytes_sent  \\\n",
            "count   10000.000000  10000.000000  10000.000000  10000.000000  10000.000000   \n",
            "mean       97.749895      8.826900     10.037581     49.910200    992.728967   \n",
            "std        97.440471      5.592465      7.131025      7.056707    975.779826   \n",
            "min         0.001163      1.000000      0.027616     27.000000      0.130020   \n",
            "25%        28.279917      6.000000      4.736240     45.000000    291.032711   \n",
            "50%        67.831496      6.000000      8.403373     50.000000    698.934620   \n",
            "75%       134.709830     17.000000     13.590517     55.000000   1385.699766   \n",
            "max       817.244560     17.000000     68.473294     75.000000   8987.990266   \n",
            "\n",
            "       bytes_received     tcp_flags  packet_size_mean  packet_size_std  \\\n",
            "count    10000.000000  10000.000000      10000.000000     10000.000000   \n",
            "mean       806.500886     13.932200        500.878838        49.696609   \n",
            "std        795.824294     11.088692        152.156913        48.921880   \n",
            "min          0.144255      0.000000        -58.064232         0.014490   \n",
            "25%        230.996580      2.000000        397.352328        14.416769   \n",
            "50%        567.059669     16.000000        502.941119        34.404307   \n",
            "75%       1135.501234     24.000000        602.771746        68.480492   \n",
            "max       7141.336086     27.000000       1024.742540       427.326478   \n",
            "\n",
            "       inter_arrival_time     flow_rate   source_port  destination_port  \n",
            "count        10000.000000  10000.000000  10000.000000      10000.000000  \n",
            "mean             0.099476     19.823241  33020.265600       1230.769200  \n",
            "std              0.100492     14.216574  18502.730656       2776.367865  \n",
            "min              0.000024      0.126125   1025.000000         22.000000  \n",
            "25%              0.028002      9.496867  16924.750000         23.000000  \n",
            "50%              0.068702     16.486388  32428.000000         53.000000  \n",
            "75%              0.138353     26.555956  48870.500000        443.000000  \n",
            "max              0.939986    159.897972  65532.000000       8080.000000  \n",
            "\n",
            "================================================================================\n",
            "PHASE 2: DATA PREPARATION\n",
            "================================================================================\n",
            "\n",
            "============================================================\n",
            "COMPLETE DATA PREPARATION PIPELINE\n",
            "============================================================\n",
            "\n",
            "============================================================\n",
            "DATA PREPARATION PHASE\n",
            "============================================================\n",
            "\n",
            "Handling Missing Values...\n",
            "Missing values before: 0\n",
            "Missing values after: 0\n",
            "\n",
            "Defining Risk Outcome...\n",
            "Risk Level Distribution:\n",
            "risk_binary\n",
            "0    9165\n",
            "1     835\n",
            "Name: count, dtype: int64\n",
            "Risk Level Percentage:\n",
            "risk_binary\n",
            "0    91.65\n",
            "1     8.35\n",
            "Name: proportion, dtype: float64\n",
            "\n",
            "Encoding Categorical Features...\n",
            "\n",
            "Splitting data into train and test sets (test_size=0.2)...\n",
            "Training set: (8000, 13), Test set: (2000, 13)\n",
            "Training class distribution: [7332  668]\n",
            "Test class distribution: [1833  167]\n",
            "\n",
            "Feature Selection...\n",
            "\n",
            "Feature Importance Scores:\n",
            "               feature     score\n",
            "6            tcp_flags  0.006749\n",
            "0        flow_duration  0.004673\n",
            "3         packet_count  0.003342\n",
            "4           bytes_sent  0.003269\n",
            "7     packet_size_mean  0.003032\n",
            "8      packet_size_std  0.001723\n",
            "10           flow_rate  0.001077\n",
            "1             protocol  0.000983\n",
            "9   inter_arrival_time  0.000254\n",
            "2          packet_rate  0.000000\n",
            "\n",
            "Selected 7 features (keeping 95% of importance)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x1000 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Scaling Features...\n",
            "Features scaled. Training shape: (8000, 7), Test shape: (2000, 7)\n",
            "\n",
            "Handling Class Imbalance...\n",
            "Before SMOTE - Class distribution in training: [7332  668]\n",
            "After SMOTE - Class distribution in training: [7332 7332]\n",
            "\n",
            "================================================================================\n",
            "PHASE 3: MODEL DEVELOPMENT\n",
            "================================================================================\n",
            "\n",
            "============================================================\n",
            "TRAINING LOGISTIC REGRESSION MODEL\n",
            "============================================================\n",
            "Performing hyperparameter tuning with GridSearchCV...\n",
            "Best parameters: {'C': 0.01, 'max_iter': 1000, 'solver': 'liblinear'}\n",
            "Best CV F1 score: 0.5592\n",
            "\n",
            "============================================================\n",
            "TRAINING RANDOM FOREST MODEL\n",
            "============================================================\n",
            "Performing hyperparameter tuning with GridSearchCV (this may take a while)...\n",
            "Best parameters: {'max_depth': 20, 'min_samples_split': 2, 'n_estimators': 200}\n",
            "Best CV F1 score: 0.9451\n",
            "\n",
            "Top 10 Feature Importances:\n",
            "            feature  importance\n",
            "0         tcp_flags    0.250124\n",
            "2      packet_count    0.151325\n",
            "3        bytes_sent    0.123074\n",
            "4  packet_size_mean    0.121770\n",
            "6         flow_rate    0.120636\n",
            "1     flow_duration    0.118523\n",
            "5   packet_size_std    0.114548\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x800 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "============================================================\n",
            "MODEL EVALUATION RESULTS\n",
            "============================================================\n",
            "\n",
            "==================================================\n",
            "EVALUATING LOGISTIC_REGRESSION\n",
            "==================================================\n",
            "Accuracy:  0.4810\n",
            "Precision: 0.0895\n",
            "Recall:    0.5689\n",
            "F1 Score:  0.1547\n",
            "ROC-AUC:   0.5436\n",
            "\n",
            "Confusion Matrix:\n",
            "          Predicted\n",
            "          Low Risk  High Risk\n",
            "Low Risk    867        966\n",
            "High Risk   72         95\n",
            "\n",
            "Detailed Classification Report:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "    Low Risk       0.92      0.47      0.63      1833\n",
            "   High Risk       0.09      0.57      0.15       167\n",
            "\n",
            "    accuracy                           0.48      2000\n",
            "   macro avg       0.51      0.52      0.39      2000\n",
            "weighted avg       0.85      0.48      0.59      2000\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1400x500 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "==================================================\n",
            "EVALUATING RANDOM_FOREST\n",
            "==================================================\n",
            "Accuracy:  0.8935\n",
            "Precision: 0.1515\n",
            "Recall:    0.0599\n",
            "F1 Score:  0.0858\n",
            "ROC-AUC:   0.5310\n",
            "\n",
            "Confusion Matrix:\n",
            "          Predicted\n",
            "          Low Risk  High Risk\n",
            "Low Risk    1777       56\n",
            "High Risk   157        10\n",
            "\n",
            "Detailed Classification Report:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "    Low Risk       0.92      0.97      0.94      1833\n",
            "   High Risk       0.15      0.06      0.09       167\n",
            "\n",
            "    accuracy                           0.89      2000\n",
            "   macro avg       0.54      0.51      0.51      2000\n",
            "weighted avg       0.85      0.89      0.87      2000\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1400x500 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "============================================================\n",
            "MODEL COMPARISON\n",
            "============================================================\n",
            "\n",
            "Performance Comparison Table:\n",
            "                    accuracy precision    recall  f1_score   roc_auc\n",
            "logistic_regression    0.481  0.089538  0.568862  0.154723  0.543636\n",
            "random_forest         0.8935  0.151515   0.05988  0.085837  0.530981\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x800 with 6 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "================================================================================\n",
            "PHASE 4: COMPARATIVE ASSESSMENT\n",
            "================================================================================\n",
            "\n",
            "============================================================\n",
            "CONVENTIONAL RISK ASSESSMENT APPROACH\n",
            "============================================================\n",
            "Conventional Assessment Results:\n",
            "Accuracy:  0.7350\n",
            "Precision: 0.0689\n",
            "Recall:    0.1737\n",
            "F1 Score:  0.0986\n",
            "\n",
            "============================================================\n",
            "COMPARATIVE ANALYSIS: ML vs CONVENTIONAL APPROACH\n",
            "============================================================\n",
            "\n",
            "Comparison Table:\n",
            "                     accuracy  precision  recall  f1_score\n",
            "Conventional           0.7350     0.0689  0.1737    0.0986\n",
            "logistic_regression    0.4810     0.0895  0.5689    0.1547\n",
            "random_forest          0.8935     0.1515  0.0599    0.0858\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x1000 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Improvement of ML over Conventional Approach:\n",
            "\n",
            "LOGISTIC_REGRESSION:\n",
            "  accuracy: -34.6% improvement\n",
            "  precision: 30.0% improvement\n",
            "  recall: 227.6% improvement\n",
            "  f1_score: 56.9% improvement\n",
            "\n",
            "RANDOM_FOREST:\n",
            "  accuracy: 21.6% improvement\n",
            "  precision: 120.0% improvement\n",
            "  recall: -65.5% improvement\n",
            "  f1_score: -13.0% improvement\n",
            "\n",
            "================================================================================\n",
            "EXECUTION COMPLETE\n",
            "================================================================================\n",
            "\n",
            "The machine learning-based quantitative cybersecurity risk assessment\n",
            "model has been successfully developed and evaluated.\n",
            "\n",
            "Key Findings:\n",
            "Best Model: logistic_regression\n",
            "Best F1 Score: 0.1547\n",
            "\n",
            "================================================================================\n",
            "INTERPRETATION IN CONTEXT OF RESEARCH\n",
            "================================================================================\n",
            "\n",
            "This implementation demonstrates:\n",
            "1. Data understanding and preparation using CRISP-DM framework\n",
            "2. Risk outcome operationalization for IoT cybersecurity\n",
            "3. Machine learning models for quantitative risk assessment\n",
            "4. Evaluation using Accuracy, Precision, Recall, F1-score, and ROC-AUC\n",
            "5. Comparative assessment with conventional approach\n",
            "\n",
            "The results provide evidence for quantitative risk assessment\n",
            "and prioritization in heterogeneous IoT environments.\n"
          ]
        }
      ],
      "source": [
        "\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from sklearn.model_selection import train_test_split, cross_val_score, GridSearchCV\n",
        "from sklearn.preprocessing import StandardScaler, LabelEncoder, RobustScaler\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import (accuracy_score, precision_score, recall_score,\n",
        "                            f1_score, roc_auc_score, confusion_matrix,\n",
        "                            classification_report, roc_curve)\n",
        "from sklearn.feature_selection import SelectKBest, f_classif, mutual_info_classif\n",
        "from imblearn.over_sampling import SMOTE\n",
        "from imblearn.pipeline import Pipeline as ImbPipeline\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "\n",
        "# Set random seed for reproducibility\n",
        "RANDOM_SEED = 42\n",
        "np.random.seed(RANDOM_SEED)\n",
        "\n",
        "# =============================================================================\n",
        "# Step 1: Data Understanding (CRISP-DM Phase 2)\n",
        "# =============================================================================\n",
        "\n",
        "class DataUnderstanding:\n",
        "    \"\"\"\n",
        "    Data Understanding phase as per CRISP-DM methodology.\n",
        "    Examines data structure, variables, attack categories, and class distribution.\n",
        "    \"\"\"\n",
        "\n",
        "    def __init__(self, file_path):\n",
        "        self.file_path = file_path\n",
        "        self.data = None\n",
        "        self.data_summary = {}\n",
        "\n",
        "    def load_data(self):\n",
        "        \"\"\"Load the CICIoT2023 dataset\"\"\"\n",
        "        # Note: Replace with actual file path\n",
        "        # self.data = pd.read_csv(self.file_path)\n",
        "        # For demonstration, we'll create a synthetic dataset structure\n",
        "        print(\"Loading CICIoT2023 dataset...\")\n",
        "        print(\"Note: Using synthetic data for demonstration. Replace with actual CICIoT2023 dataset.\")\n",
        "        self.data = self._create_demo_dataset()\n",
        "        return self.data\n",
        "\n",
        "    def _create_demo_dataset(self):\n",
        "        \"\"\"Create a synthetic dataset mimicking CICIoT2023 structure\"\"\"\n",
        "        np.random.seed(RANDOM_SEED)\n",
        "        n_samples = 10000\n",
        "\n",
        "        # Simulate network flow features based on CICIoT2023 characteristics\n",
        "        data = {\n",
        "            'flow_duration': np.random.exponential(100, n_samples),\n",
        "            'protocol': np.random.choice([6, 17, 1], n_samples, p=[0.6, 0.3, 0.1]),  # TCP, UDP, ICMP\n",
        "            'packet_rate': np.random.gamma(2, 5, n_samples),\n",
        "            'packet_count': np.random.poisson(50, n_samples),\n",
        "            'bytes_sent': np.random.exponential(1000, n_samples),\n",
        "            'bytes_received': np.random.exponential(800, n_samples),\n",
        "            'tcp_flags': np.random.choice([0, 2, 16, 24, 27], n_samples),\n",
        "            'packet_size_mean': np.random.normal(500, 150, n_samples),\n",
        "            'packet_size_std': np.random.exponential(50, n_samples),\n",
        "            'inter_arrival_time': np.random.exponential(0.1, n_samples),\n",
        "            'flow_rate': np.random.gamma(2, 10, n_samples),\n",
        "            'source_port': np.random.randint(1024, 65535, n_samples),\n",
        "            'destination_port': np.random.choice([80, 443, 22, 53, 25, 23, 8080], n_samples),\n",
        "        }\n",
        "\n",
        "        df = pd.DataFrame(data)\n",
        "\n",
        "        # Create synthetic attack categories based on feature combinations\n",
        "        attack_prob = np.zeros(n_samples)\n",
        "\n",
        "        # DDoS patterns\n",
        "        ddos_idx = np.random.choice(n_samples, size=int(n_samples*0.15), replace=False)\n",
        "        attack_prob[ddos_idx] += 3\n",
        "\n",
        "        # DoS patterns\n",
        "        dos_idx = np.random.choice(n_samples, size=int(n_samples*0.10), replace=False)\n",
        "        attack_prob[dos_idx] += 2\n",
        "\n",
        "        # Reconnaissance patterns\n",
        "        recon_idx = np.random.choice(n_samples, size=int(n_samples*0.08), replace=False)\n",
        "        attack_prob[recon_idx] += 1.5\n",
        "\n",
        "        # Brute Force patterns\n",
        "        brute_idx = np.random.choice(n_samples, size=int(n_samples*0.05), replace=False)\n",
        "        attack_prob[brute_idx] += 1\n",
        "\n",
        "        # Assign labels based on attack probability\n",
        "        df['attack_category'] = 'Benign'\n",
        "        attack_threshold = 2.0\n",
        "\n",
        "        # Mark attacks\n",
        "        attack_idx = np.where(attack_prob >= attack_threshold)[0]\n",
        "        df.loc[attack_idx, 'attack_category'] = 'Malicious'\n",
        "\n",
        "        # Further categorize attacks\n",
        "        for idx in attack_idx[:1000]:  # Assign specific attack types to some samples\n",
        "            if attack_prob[idx] >= 3.5:\n",
        "                df.loc[idx, 'attack_category'] = 'DDoS'\n",
        "            elif attack_prob[idx] >= 2.8:\n",
        "                df.loc[idx, 'attack_category'] = 'DoS'\n",
        "            elif attack_prob[idx] >= 2.2:\n",
        "                df.loc[idx, 'attack_category'] = 'Reconnaissance'\n",
        "            elif attack_prob[idx] >= 1.8:\n",
        "                df.loc[idx, 'attack_category'] = 'Brute Force'\n",
        "\n",
        "        # Add some spoofing and Mirai samples\n",
        "        spoof_idx = np.random.choice(n_samples, size=int(n_samples*0.03), replace=False)\n",
        "        df.loc[spoof_idx, 'attack_category'] = 'Spoofing'\n",
        "\n",
        "        mirai_idx = np.random.choice(n_samples, size=int(n_samples*0.02), replace=False)\n",
        "        df.loc[mirai_idx, 'attack_category'] = 'Mirai'\n",
        "\n",
        "        # Add some noise\n",
        "        df['attack_category'] = df['attack_category'].astype('category')\n",
        "\n",
        "        return df\n",
        "\n",
        "    def explore_data(self):\n",
        "        \"\"\"Perform exploratory data analysis\"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"DATA UNDERSTANDING PHASE\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        print(f\"\\nDataset Shape: {self.data.shape}\")\n",
        "        print(f\"\\nColumns: {list(self.data.columns)}\")\n",
        "\n",
        "        print(\"\\nData Types:\")\n",
        "        print(self.data.dtypes)\n",
        "\n",
        "        print(\"\\nMissing Values:\")\n",
        "        print(self.data.isnull().sum())\n",
        "\n",
        "        print(\"\\nAttack Category Distribution:\")\n",
        "        print(self.data['attack_category'].value_counts())\n",
        "        print(f\"\\nPercentage Distribution:\")\n",
        "        print(self.data['attack_category'].value_counts(normalize=True) * 100)\n",
        "\n",
        "        print(\"\\nNumerical Features Summary:\")\n",
        "        print(self.data.describe())\n",
        "\n",
        "        return self.data\n",
        "\n",
        "    def visualize_distributions(self):\n",
        "        \"\"\"Visualize data distributions\"\"\"\n",
        "        fig, axes = plt.subplots(3, 4, figsize=(16, 12))\n",
        "        axes = axes.flatten()\n",
        "\n",
        "        numeric_cols = self.data.select_dtypes(include=[np.number]).columns[:12]\n",
        "\n",
        "        for idx, col in enumerate(numeric_cols):\n",
        "            self.data[col].hist(bins=30, ax=axes[idx], alpha=0.7)\n",
        "            axes[idx].set_title(f'Distribution of {col}')\n",
        "            axes[idx].set_xlabel(col)\n",
        "            axes[idx].set_ylabel('Frequency')\n",
        "\n",
        "        plt.tight_layout()\n",
        "        plt.savefig('data_distributions.png', dpi=300, bbox_inches='tight')\n",
        "        plt.show()\n",
        "\n",
        "        # Attack category distribution\n",
        "        fig, ax = plt.subplots(figsize=(10, 6))\n",
        "        self.data['attack_category'].value_counts().plot(kind='bar', ax=ax)\n",
        "        ax.set_title('Attack Category Distribution')\n",
        "        ax.set_xlabel('Attack Category')\n",
        "        ax.set_ylabel('Count')\n",
        "        plt.xticks(rotation=45)\n",
        "        plt.tight_layout()\n",
        "        plt.savefig('attack_distribution.png', dpi=300, bbox_inches='tight')\n",
        "        plt.show()\n",
        "\n",
        "# =============================================================================\n",
        "# Step 2: Data Preparation (CRISP-DM Phase 3)\n",
        "# =============================================================================\n",
        "\n",
        "class DataPreparation:\n",
        "    \"\"\"\n",
        "    Data Preparation phase as per CRISP-DM methodology.\n",
        "    Handles missing values, encoding, scaling, and class imbalance.\n",
        "    \"\"\"\n",
        "\n",
        "    def __init__(self, data):\n",
        "        self.data = data.copy()\n",
        "        self.numeric_features = []\n",
        "        self.categorical_features = []\n",
        "        self.target_col = 'risk_level'  # Will be created\n",
        "        self.X_train = None\n",
        "        self.X_test = None\n",
        "        self.y_train = None\n",
        "        self.y_test = None\n",
        "\n",
        "    def handle_missing_values(self):\n",
        "        \"\"\"Handle missing values in the dataset\"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"DATA PREPARATION PHASE\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        print(\"\\nHandling Missing Values...\")\n",
        "        missing_before = self.data.isnull().sum().sum()\n",
        "\n",
        "        # For numerical columns, fill with median\n",
        "        numeric_cols = self.data.select_dtypes(include=[np.number]).columns\n",
        "        for col in numeric_cols:\n",
        "            if self.data[col].isnull().sum() > 0:\n",
        "                self.data[col].fillna(self.data[col].median(), inplace=True)\n",
        "\n",
        "        # For categorical columns, fill with mode\n",
        "        categorical_cols = self.data.select_dtypes(include=['category', 'object']).columns\n",
        "        for col in categorical_cols:\n",
        "            if self.data[col].isnull().sum() > 0:\n",
        "                self.data[col].fillna(self.data[col].mode()[0], inplace=True)\n",
        "\n",
        "        missing_after = self.data.isnull().sum().sum()\n",
        "        print(f\"Missing values before: {missing_before}\")\n",
        "        print(f\"Missing values after: {missing_after}\")\n",
        "\n",
        "        return self.data\n",
        "\n",
        "    def define_risk_outcome(self):\n",
        "        \"\"\"\n",
        "        Operationalize the cybersecurity risk outcome (3.10)\n",
        "        Creates binary classification: low risk vs high risk\n",
        "        \"\"\"\n",
        "        print(\"\\nDefining Risk Outcome...\")\n",
        "\n",
        "        # Map attack categories to risk levels\n",
        "        risk_mapping = {\n",
        "            'Benign': 0,  # Low risk\n",
        "            'Reconnaissance': 1,  # Medium risk\n",
        "            'Brute Force': 1,\n",
        "            'Spoofing': 1,\n",
        "            'DoS': 2,  # High risk\n",
        "            'DDoS': 2,\n",
        "            'Mirai': 2\n",
        "        }\n",
        "\n",
        "        self.data['risk_level'] = self.data['attack_category'].map(risk_mapping)\n",
        "\n",
        "        # Convert to binary: High risk (1) vs Low risk (0)\n",
        "        self.data['risk_binary'] = (self.data['risk_level'] >= 2).astype(int)\n",
        "\n",
        "        print(f\"Risk Level Distribution:\")\n",
        "        print(self.data['risk_binary'].value_counts())\n",
        "        print(f\"Risk Level Percentage:\")\n",
        "        print(self.data['risk_binary'].value_counts(normalize=True) * 100)\n",
        "\n",
        "        return self.data\n",
        "\n",
        "    def encode_categorical_features(self):\n",
        "        \"\"\"Encode categorical features for machine learning\"\"\"\n",
        "        print(\"\\nEncoding Categorical Features...\")\n",
        "\n",
        "        self.categorical_features = self.data.select_dtypes(include=['category', 'object']).columns\n",
        "        self.categorical_features = [col for col in self.categorical_features if col not in ['attack_category', 'risk_level', 'risk_binary']]\n",
        "\n",
        "        for col in self.categorical_features:\n",
        "            le = LabelEncoder()\n",
        "            self.data[col] = le.fit_transform(self.data[col].astype(str))\n",
        "            print(f\"Encoded {col}: {len(le.classes_)} unique values\")\n",
        "\n",
        "        return self.data\n",
        "\n",
        "    def select_features(self, X, y):\n",
        "        \"\"\"\n",
        "        Feature selection based on literature review and data characteristics\n",
        "        Identifies relevant cybersecurity risk indicators\n",
        "        \"\"\"\n",
        "        print(\"\\nFeature Selection...\")\n",
        "\n",
        "        # Based on Chapter 3.9: Risk indicators include:\n",
        "        # network-flow behavior, protocol-related characteristics,\n",
        "        # packet transmission rates, packet counts, connection duration\n",
        "\n",
        "        # Remove low-value columns and potential leakage\n",
        "        columns_to_drop = ['attack_category', 'risk_level']\n",
        "        X_selected = X.drop(columns=columns_to_drop, errors='ignore')\n",
        "\n",
        "        # Feature importance using mutual information\n",
        "        selector = SelectKBest(mutual_info_classif, k='all')\n",
        "        selector.fit(X_selected, y)\n",
        "\n",
        "        # Get feature scores\n",
        "        feature_scores = pd.DataFrame({\n",
        "            'feature': X_selected.columns,\n",
        "            'score': selector.scores_\n",
        "        }).sort_values('score', ascending=False)\n",
        "\n",
        "        print(\"\\nFeature Importance Scores:\")\n",
        "        print(feature_scores.head(10))\n",
        "\n",
        "        # Select top features (keep at least 80% of cumulative importance)\n",
        "        cumulative_importance = feature_scores['score'].cumsum() / feature_scores['score'].sum()\n",
        "        n_features = (cumulative_importance <= 0.95).sum() + 1\n",
        "        top_features = feature_scores.head(n_features)['feature'].tolist()\n",
        "\n",
        "        print(f\"\\nSelected {len(top_features)} features (keeping 95% of importance)\")\n",
        "\n",
        "        # Create correlation matrix for selected features\n",
        "        plt.figure(figsize=(12, 10))\n",
        "        correlation_matrix = X_selected[top_features].corr()\n",
        "        sns.heatmap(correlation_matrix, annot=False, cmap='coolwarm', center=0)\n",
        "        plt.title('Feature Correlation Matrix')\n",
        "        plt.tight_layout()\n",
        "        plt.savefig('feature_correlation.png', dpi=300, bbox_inches='tight')\n",
        "        plt.show()\n",
        "\n",
        "        return X_selected[top_features]\n",
        "\n",
        "    def handle_class_imbalance(self, X_train, y_train, X_test, y_test):\n",
        "        \"\"\"\n",
        "        Handle class imbalance using SMOTE (3.9)\n",
        "        Only applied to training set to prevent data leakage\n",
        "        \"\"\"\n",
        "        print(\"\\nHandling Class Imbalance...\")\n",
        "        print(f\"Before SMOTE - Class distribution in training: {np.bincount(y_train)}\")\n",
        "\n",
        "        # Apply SMOTE only on training data\n",
        "        smote = SMOTE(random_state=RANDOM_SEED)\n",
        "        X_train_resampled, y_train_resampled = smote.fit_resample(X_train, y_train)\n",
        "\n",
        "        print(f\"After SMOTE - Class distribution in training: {np.bincount(y_train_resampled)}\")\n",
        "\n",
        "        return X_train_resampled, y_train_resampled\n",
        "\n",
        "    def scale_features(self, X_train, X_test):\n",
        "        \"\"\"\n",
        "        Scale features using parameters from training data only\n",
        "        Prevents information leakage\n",
        "        \"\"\"\n",
        "        print(\"\\nScaling Features...\")\n",
        "\n",
        "        scaler = StandardScaler()\n",
        "        X_train_scaled = scaler.fit_transform(X_train)\n",
        "        X_test_scaled = scaler.transform(X_test)\n",
        "\n",
        "        # Convert back to DataFrame for interpretability\n",
        "        X_train_scaled = pd.DataFrame(X_train_scaled, columns=X_train.columns, index=X_train.index)\n",
        "        X_test_scaled = pd.DataFrame(X_test_scaled, columns=X_test.columns, index=X_test.index)\n",
        "\n",
        "        print(f\"Features scaled. Training shape: {X_train_scaled.shape}, Test shape: {X_test_scaled.shape}\")\n",
        "\n",
        "        return X_train_scaled, X_test_scaled, scaler\n",
        "\n",
        "    def prepare_data(self, test_size=0.2, use_smote=True):\n",
        "        \"\"\"\n",
        "        Complete data preparation pipeline\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"COMPLETE DATA PREPARATION PIPELINE\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        # Step 1: Handle missing values\n",
        "        self.handle_missing_values()\n",
        "\n",
        "        # Step 2: Define risk outcome\n",
        "        self.define_risk_outcome()\n",
        "\n",
        "        # Step 3: Encode categorical features\n",
        "        self.encode_categorical_features()\n",
        "\n",
        "        # Step 4: Prepare X and y\n",
        "        X = self.data.drop(['risk_binary', 'risk_level', 'attack_category'], axis=1, errors='ignore')\n",
        "        y = self.data['risk_binary']\n",
        "\n",
        "        # Step 5: Split data (3.11 - Model Training, Testing and Evaluation)\n",
        "        print(f\"\\nSplitting data into train and test sets (test_size={test_size})...\")\n",
        "        X_train, X_test, y_train, y_test = train_test_split(\n",
        "            X, y, test_size=test_size, random_state=RANDOM_SEED, stratify=y\n",
        "        )\n",
        "\n",
        "        print(f\"Training set: {X_train.shape}, Test set: {X_test.shape}\")\n",
        "        print(f\"Training class distribution: {np.bincount(y_train)}\")\n",
        "        print(f\"Test class distribution: {np.bincount(y_test)}\")\n",
        "\n",
        "        # Step 6: Feature selection (on training data only)\n",
        "        X_train_selected = self.select_features(X_train, y_train)\n",
        "        X_test_selected = X_test[X_train_selected.columns]\n",
        "\n",
        "        # Step 7: Scale features (3.9)\n",
        "        X_train_scaled, X_test_scaled, scaler = self.scale_features(X_train_selected, X_test_selected)\n",
        "\n",
        "        # Step 8: Handle class imbalance (3.9)\n",
        "        if use_smote:\n",
        "            X_train_scaled, y_train = self.handle_class_imbalance(\n",
        "                X_train_scaled, y_train, X_test_scaled, y_test\n",
        "            )\n",
        "\n",
        "        self.X_train = X_train_scaled\n",
        "        self.X_test = X_test_scaled\n",
        "        self.y_train = y_train\n",
        "        self.y_test = y_test\n",
        "        self.scaler = scaler\n",
        "\n",
        "        return self.X_train, self.X_test, self.y_train, self.y_test\n",
        "\n",
        "# =============================================================================\n",
        "# Step 3: Modeling (CRISP-DM Phase 4)\n",
        "# =============================================================================\n",
        "\n",
        "class ModelDevelopment:\n",
        "    \"\"\"\n",
        "    Model Development phase as per CRISP-DM methodology.\n",
        "    Implements Logistic Regression and Random Forest models (3.10)\n",
        "    \"\"\"\n",
        "\n",
        "    def __init__(self, X_train, X_test, y_train, y_test):\n",
        "        self.X_train = X_train\n",
        "        self.X_test = X_test\n",
        "        self.y_train = y_train\n",
        "        self.y_test = y_test\n",
        "        self.models = {}\n",
        "        self.results = {}\n",
        "\n",
        "    def train_logistic_regression(self):\n",
        "        \"\"\"\n",
        "        Train Logistic Regression baseline model (3.10)\n",
        "        Interpretable model for binary outcome\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"TRAINING LOGISTIC REGRESSION MODEL\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        # Hyperparameter tuning using GridSearchCV\n",
        "        param_grid = {\n",
        "            'C': [0.01, 0.1, 1, 10, 100],\n",
        "            'solver': ['liblinear', 'lbfgs'],\n",
        "            'max_iter': [1000, 2000]\n",
        "        }\n",
        "\n",
        "        lr = LogisticRegression(random_state=RANDOM_SEED, class_weight='balanced')\n",
        "\n",
        "        print(\"Performing hyperparameter tuning with GridSearchCV...\")\n",
        "        grid_search = GridSearchCV(\n",
        "            lr, param_grid, cv=5, scoring='f1', n_jobs=-1, verbose=0\n",
        "        )\n",
        "        grid_search.fit(self.X_train, self.y_train)\n",
        "\n",
        "        self.models['logistic_regression'] = grid_search.best_estimator_\n",
        "        print(f\"Best parameters: {grid_search.best_params_}\")\n",
        "        print(f\"Best CV F1 score: {grid_search.best_score_:.4f}\")\n",
        "\n",
        "        return self.models['logistic_regression']\n",
        "\n",
        "    def train_random_forest(self):\n",
        "        \"\"\"\n",
        "        Train Random Forest model (3.10)\n",
        "        Captures nonlinear relationships and provides feature importance\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"TRAINING RANDOM FOREST MODEL\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        # Hyperparameter tuning\n",
        "        param_grid = {\n",
        "            'n_estimators': [100, 200, 300],\n",
        "            'max_depth': [10, 20, 30, None],\n",
        "            'min_samples_split': [2, 5, 10],\n",
        "            'min_samples_leaf': [1, 2, 4],\n",
        "            'max_features': ['sqrt', 'log2', None]\n",
        "        }\n",
        "\n",
        "        rf = RandomForestClassifier(\n",
        "            random_state=RANDOM_SEED,\n",
        "            class_weight='balanced',\n",
        "            n_jobs=-1\n",
        "        )\n",
        "\n",
        "        print(\"Performing hyperparameter tuning with GridSearchCV (this may take a while)...\")\n",
        "        # Use a smaller grid for speed\n",
        "        param_grid_small = {\n",
        "            'n_estimators': [100, 200],\n",
        "            'max_depth': [10, 20],\n",
        "            'min_samples_split': [2, 5]\n",
        "        }\n",
        "\n",
        "        grid_search = GridSearchCV(\n",
        "            rf, param_grid_small, cv=5, scoring='f1', n_jobs=-1, verbose=0\n",
        "        )\n",
        "        grid_search.fit(self.X_train, self.y_train)\n",
        "\n",
        "        self.models['random_forest'] = grid_search.best_estimator_\n",
        "        print(f\"Best parameters: {grid_search.best_params_}\")\n",
        "        print(f\"Best CV F1 score: {grid_search.best_score_:.4f}\")\n",
        "\n",
        "        # Feature importance\n",
        "        feature_importance = pd.DataFrame({\n",
        "            'feature': self.X_train.columns,\n",
        "            'importance': self.models['random_forest'].feature_importances_\n",
        "        }).sort_values('importance', ascending=False)\n",
        "\n",
        "        print(\"\\nTop 10 Feature Importances:\")\n",
        "        print(feature_importance.head(10))\n",
        "\n",
        "        # Visualize feature importance\n",
        "        plt.figure(figsize=(10, 8))\n",
        "        plt.barh(feature_importance['feature'][:15][::-1],\n",
        "                feature_importance['importance'][:15][::-1])\n",
        "        plt.xlabel('Feature Importance')\n",
        "        plt.title('Top 15 Feature Importances - Random Forest')\n",
        "        plt.tight_layout()\n",
        "        plt.savefig('feature_importance.png', dpi=300, bbox_inches='tight')\n",
        "        plt.show()\n",
        "\n",
        "        return self.models['random_forest']\n",
        "\n",
        "    def evaluate_models(self):\n",
        "        \"\"\"\n",
        "        Evaluate all trained models (3.11)\n",
        "        Uses Accuracy, Precision, Recall, F1-score, ROC-AUC\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"MODEL EVALUATION RESULTS\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        for model_name, model in self.models.items():\n",
        "            print(f\"\\n{'='*50}\")\n",
        "            print(f\"EVALUATING {model_name.upper()}\")\n",
        "            print(f\"{'='*50}\")\n",
        "\n",
        "            # Predictions\n",
        "            y_pred = model.predict(self.X_test)\n",
        "            y_pred_proba = model.predict_proba(self.X_test)[:, 1]\n",
        "\n",
        "            # Calculate metrics\n",
        "            accuracy = accuracy_score(self.y_test, y_pred)\n",
        "            precision = precision_score(self.y_test, y_pred)\n",
        "            recall = recall_score(self.y_test, y_pred)\n",
        "            f1 = f1_score(self.y_test, y_pred)\n",
        "            roc_auc = roc_auc_score(self.y_test, y_pred_proba)\n",
        "\n",
        "            self.results[model_name] = {\n",
        "                'accuracy': accuracy,\n",
        "                'precision': precision,\n",
        "                'recall': recall,\n",
        "                'f1_score': f1,\n",
        "                'roc_auc': roc_auc,\n",
        "                'y_pred': y_pred,\n",
        "                'y_pred_proba': y_pred_proba\n",
        "            }\n",
        "\n",
        "            print(f\"Accuracy:  {accuracy:.4f}\")\n",
        "            print(f\"Precision: {precision:.4f}\")\n",
        "            print(f\"Recall:    {recall:.4f}\")\n",
        "            print(f\"F1 Score:  {f1:.4f}\")\n",
        "            print(f\"ROC-AUC:   {roc_auc:.4f}\")\n",
        "\n",
        "            # Confusion Matrix\n",
        "            cm = confusion_matrix(self.y_test, y_pred)\n",
        "            print(f\"\\nConfusion Matrix:\")\n",
        "            print(f\"          Predicted\")\n",
        "            print(f\"          Low Risk  High Risk\")\n",
        "            print(f\"Low Risk    {cm[0][0]:<10} {cm[0][1]}\")\n",
        "            print(f\"High Risk   {cm[1][0]:<10} {cm[1][1]}\")\n",
        "\n",
        "            # Classification Report\n",
        "            print(f\"\\nDetailed Classification Report:\")\n",
        "            print(classification_report(self.y_test, y_pred,\n",
        "                                      target_names=['Low Risk', 'High Risk']))\n",
        "\n",
        "            # Visualize confusion matrix\n",
        "            fig, ax = plt.subplots(1, 2, figsize=(14, 5))\n",
        "\n",
        "            # Confusion matrix heatmap\n",
        "            sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', ax=ax[0])\n",
        "            ax[0].set_title(f'Confusion Matrix - {model_name.upper()}')\n",
        "            ax[0].set_xlabel('Predicted')\n",
        "            ax[0].set_ylabel('Actual')\n",
        "            ax[0].set_xticklabels(['Low Risk', 'High Risk'])\n",
        "            ax[0].set_yticklabels(['Low Risk', 'High Risk'])\n",
        "\n",
        "            # ROC Curve\n",
        "            fpr, tpr, _ = roc_curve(self.y_test, y_pred_proba)\n",
        "            ax[1].plot(fpr, tpr, label=f'ROC-AUC = {roc_auc:.4f}')\n",
        "            ax[1].plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
        "            ax[1].set_xlabel('False Positive Rate')\n",
        "            ax[1].set_ylabel('True Positive Rate')\n",
        "            ax[1].set_title(f'ROC Curve - {model_name.upper()}')\n",
        "            ax[1].legend()\n",
        "\n",
        "            plt.tight_layout()\n",
        "            plt.savefig(f'confusion_roc_{model_name}.png', dpi=300, bbox_inches='tight')\n",
        "            plt.show()\n",
        "\n",
        "        # Compare models\n",
        "        self.compare_models()\n",
        "\n",
        "        return self.results\n",
        "\n",
        "    def compare_models(self):\n",
        "        \"\"\"\n",
        "        Compare performance of all models\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"MODEL COMPARISON\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        comparison_df = pd.DataFrame(self.results).T\n",
        "        comparison_df = comparison_df.drop(['y_pred', 'y_pred_proba'], axis=1, errors='ignore')\n",
        "\n",
        "        print(\"\\nPerformance Comparison Table:\")\n",
        "        print(comparison_df.round(4))\n",
        "\n",
        "        # Bar chart comparison\n",
        "        fig, axes = plt.subplots(2, 3, figsize=(15, 8))\n",
        "        axes = axes.flatten()\n",
        "\n",
        "        metrics = ['accuracy', 'precision', 'recall', 'f1_score', 'roc_auc']\n",
        "        colors = ['#2E86AB', '#A23B72', '#F18F01', '#C73E1D']\n",
        "\n",
        "        for idx, metric in enumerate(metrics):\n",
        "            if metric in comparison_df.columns:\n",
        "                comparison_df[metric].plot(kind='bar', ax=axes[idx], color=colors[:len(comparison_df)])\n",
        "                axes[idx].set_title(f'{metric.upper()}')\n",
        "                axes[idx].set_ylabel('Score')\n",
        "                axes[idx].set_ylim(0, 1)\n",
        "                axes[idx].grid(True, alpha=0.3)\n",
        "                axes[idx].legend().remove()\n",
        "\n",
        "        # Hide empty subplot\n",
        "        axes[5].axis('off')\n",
        "\n",
        "        plt.tight_layout()\n",
        "        plt.savefig('model_comparison.png', dpi=300, bbox_inches='tight')\n",
        "        plt.show()\n",
        "\n",
        "# =============================================================================\n",
        "# Step 4: Comparative Assessment (3.11)\n",
        "# =============================================================================\n",
        "\n",
        "class ComparativeAssessment:\n",
        "    \"\"\"\n",
        "    Comparative assessment with conventional cybersecurity risk assessment approach\n",
        "    \"\"\"\n",
        "\n",
        "    def __init__(self, results, X_test, y_test):\n",
        "        self.results = results\n",
        "        self.X_test = X_test\n",
        "        self.y_test = y_test\n",
        "        self.conventional_results = {}\n",
        "\n",
        "    def conventional_risk_assessment(self):\n",
        "        \"\"\"\n",
        "        Simulate conventional cybersecurity risk assessment approach\n",
        "        Rule-based or score-based approach\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"CONVENTIONAL RISK ASSESSMENT APPROACH\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        # Simulate a conventional approach using heuristic rules\n",
        "        # Based on thresholds of network features\n",
        "\n",
        "        # Define risk score based on feature thresholds\n",
        "        risk_scores = pd.Series(0, index=self.X_test.index)\n",
        "\n",
        "        # Example rule: High packet rate and high flow rate indicates risk\n",
        "        if 'packet_rate' in self.X_test.columns:\n",
        "            risk_scores += (self.X_test['packet_rate'] > self.X_test['packet_rate'].mean()).astype(int)\n",
        "\n",
        "        if 'flow_rate' in self.X_test.columns:\n",
        "            risk_scores += (self.X_test['flow_rate'] > self.X_test['flow_rate'].mean()).astype(int)\n",
        "\n",
        "        if 'packet_count' in self.X_test.columns:\n",
        "            risk_scores += (self.X_test['packet_count'] > self.X_test['packet_count'].mean()).astype(int)\n",
        "\n",
        "        # Normalize risk scores\n",
        "        risk_scores = (risk_scores > risk_scores.median()).astype(int)\n",
        "\n",
        "        # Calculate conventional metrics\n",
        "        accuracy = accuracy_score(self.y_test, risk_scores)\n",
        "        precision = precision_score(self.y_test, risk_scores)\n",
        "        recall = recall_score(self.y_test, risk_scores)\n",
        "        f1 = f1_score(self.y_test, risk_scores)\n",
        "\n",
        "        self.conventional_results = {\n",
        "            'accuracy': accuracy,\n",
        "            'precision': precision,\n",
        "            'recall': recall,\n",
        "            'f1_score': f1\n",
        "        }\n",
        "\n",
        "        print(f\"Conventional Assessment Results:\")\n",
        "        print(f\"Accuracy:  {accuracy:.4f}\")\n",
        "        print(f\"Precision: {precision:.4f}\")\n",
        "        print(f\"Recall:    {recall:.4f}\")\n",
        "        print(f\"F1 Score:  {f1:.4f}\")\n",
        "\n",
        "        return self.conventional_results\n",
        "\n",
        "    def compare_with_conventional(self):\n",
        "        \"\"\"\n",
        "        Compare machine learning models with conventional approach (3.11)\n",
        "        \"\"\"\n",
        "        print(\"\\n\" + \"=\"*60)\n",
        "        print(\"COMPARATIVE ANALYSIS: ML vs CONVENTIONAL APPROACH\")\n",
        "        print(\"=\"*60)\n",
        "\n",
        "        # Combine results\n",
        "        comparison_dict = {'Conventional': self.conventional_results}\n",
        "\n",
        "        for model_name, results in self.results.items():\n",
        "            comparison_dict[model_name] = {\n",
        "                'accuracy': results['accuracy'],\n",
        "                'precision': results['precision'],\n",
        "                'recall': results['recall'],\n",
        "                'f1_score': results['f1_score']\n",
        "            }\n",
        "\n",
        "        comparison_df = pd.DataFrame(comparison_dict).T\n",
        "        print(\"\\nComparison Table:\")\n",
        "        print(comparison_df.round(4))\n",
        "\n",
        "        # Visualize comparison\n",
        "        fig, axes = plt.subplots(2, 2, figsize=(12, 10))\n",
        "        axes = axes.flatten()\n",
        "\n",
        "        metrics = ['accuracy', 'precision', 'recall', 'f1_score']\n",
        "\n",
        "        for idx, metric in enumerate(metrics):\n",
        "            comparison_df[metric].plot(kind='bar', ax=axes[idx], color=['#4C72B0', '#55A868', '#C44E52'])\n",
        "            axes[idx].set_title(f'{metric.upper()}')\n",
        "            axes[idx].set_ylabel('Score')\n",
        "            axes[idx].set_ylim(0, 1)\n",
        "            axes[idx].grid(True, alpha=0.3)\n",
        "            axes[idx].legend().remove()\n",
        "\n",
        "        plt.tight_layout()\n",
        "        plt.savefig('ml_vs_conventional_comparison.png', dpi=300, bbox_inches='tight')\n",
        "        plt.show()\n",
        "\n",
        "        # Calculate improvement\n",
        "        print(\"\\nImprovement of ML over Conventional Approach:\")\n",
        "        for model_name in self.results.keys():\n",
        "            print(f\"\\n{model_name.upper()}:\")\n",
        "            for metric in ['accuracy', 'precision', 'recall', 'f1_score']:\n",
        "                improvement = ((comparison_dict[model_name][metric] -\n",
        "                               comparison_dict['Conventional'][metric]) /\n",
        "                               comparison_dict['Conventional'][metric] * 100)\n",
        "                print(f\"  {metric}: {improvement:.1f}% improvement\")\n",
        "\n",
        "        return comparison_df\n",
        "\n",
        "# =============================================================================\n",
        "# Main Execution Pipeline\n",
        "# =============================================================================\n",
        "\n",
        "def main():\n",
        "    \"\"\"\n",
        "    Main execution pipeline following the research methodology\n",
        "    \"\"\"\n",
        "    print(\"=\"*80)\n",
        "    print(\"QUANTITATIVE CYBERSECURITY RISK ASSESSMENT MODEL\")\n",
        "    print(\"FOR HETEROGENEOUS IOT ENVIRONMENTS\")\n",
        "    print(\"Implementation of Chapter 3 Methodology\")\n",
        "    print(\"=\"*80)\n",
        "\n",
        "    # =====================================================================\n",
        "    # Phase 1: Data Understanding (CRISP-DM Phase 2)\n",
        "    # =====================================================================\n",
        "    print(\"\\n\" + \"=\"*80)\n",
        "    print(\"PHASE 1: DATA UNDERSTANDING\")\n",
        "    print(\"=\"*80)\n",
        "\n",
        "    data_understanding = DataUnderstanding('ciciot2023.csv')\n",
        "    data = data_understanding.load_data()\n",
        "    data_understanding.explore_data()\n",
        "    # data_understanding.visualize_distributions()\n",
        "\n",
        "    # =====================================================================\n",
        "    # Phase 2: Data Preparation (CRISP-DM Phase 3)\n",
        "    # =====================================================================\n",
        "    print(\"\\n\" + \"=\"*80)\n",
        "    print(\"PHASE 2: DATA PREPARATION\")\n",
        "    print(\"=\"*80)\n",
        "\n",
        "    data_preparation = DataPreparation(data)\n",
        "    X_train, X_test, y_train, y_test = data_preparation.prepare_data(test_size=0.2, use_smote=True)\n",
        "\n",
        "    # =====================================================================\n",
        "    # Phase 3: Modeling (CRISP-DM Phase 4)\n",
        "    # =====================================================================\n",
        "    print(\"\\n\" + \"=\"*80)\n",
        "    print(\"PHASE 3: MODEL DEVELOPMENT\")\n",
        "    print(\"=\"*80)\n",
        "\n",
        "    model_development = ModelDevelopment(X_train, X_test, y_train, y_test)\n",
        "\n",
        "    # Train Logistic Regression (baseline model)\n",
        "    lr_model = model_development.train_logistic_regression()\n",
        "\n",
        "    # Train Random Forest (tree-based model)\n",
        "    rf_model = model_development.train_random_forest()\n",
        "\n",
        "    # Evaluate models\n",
        "    results = model_development.evaluate_models()\n",
        "\n",
        "    # =====================================================================\n",
        "    # Phase 4: Comparative Assessment (3.11)\n",
        "    # =====================================================================\n",
        "    print(\"\\n\" + \"=\"*80)\n",
        "    print(\"PHASE 4: COMPARATIVE ASSESSMENT\")\n",
        "    print(\"=\"*80)\n",
        "\n",
        "    comparative_assessment = ComparativeAssessment(results, X_test, y_test)\n",
        "    comparative_assessment.conventional_risk_assessment()\n",
        "    comparison_df = comparative_assessment.compare_with_conventional()\n",
        "\n",
        "    # =====================================================================\n",
        "    # Summary\n",
        "    # =====================================================================\n",
        "    print(\"\\n\" + \"=\"*80)\n",
        "    print(\"EXECUTION COMPLETE\")\n",
        "    print(\"=\"*80)\n",
        "    print(\"\\nThe machine learning-based quantitative cybersecurity risk assessment\")\n",
        "    print(\"model has been successfully developed and evaluated.\")\n",
        "    print(\"\\nKey Findings:\")\n",
        "    print(f\"Best Model: {comparison_df.idxmax()['f1_score']}\")\n",
        "    print(f\"Best F1 Score: {comparison_df['f1_score'].max():.4f}\")\n",
        "\n",
        "    # Interpret results in context of research\n",
        "    print(\"\\n\" + \"=\"*80)\n",
        "    print(\"INTERPRETATION IN CONTEXT OF RESEARCH\")\n",
        "    print(\"=\"*80)\n",
        "    print(\"\\nThis implementation demonstrates:\")\n",
        "    print(\"1. Data understanding and preparation using CRISP-DM framework\")\n",
        "    print(\"2. Risk outcome operationalization for IoT cybersecurity\")\n",
        "    print(\"3. Machine learning models for quantitative risk assessment\")\n",
        "    print(\"4. Evaluation using Accuracy, Precision, Recall, F1-score, and ROC-AUC\")\n",
        "    print(\"5. Comparative assessment with conventional approach\")\n",
        "    print(\"\\nThe results provide evidence for quantitative risk assessment\")\n",
        "    print(\"and prioritization in heterogeneous IoT environments.\")\n",
        "\n",
        "# =============================================================================\n",
        "# Additional Utility Functions\n",
        "# =============================================================================\n",
        "\n",
        "def load_ciciot2023_actual(file_path):\n",
        "    \"\"\"\n",
        "    Function to load the actual CICIoT2023 dataset\n",
        "    Replace the demo data with actual data loading\n",
        "    \"\"\"\n",
        "    # Uncomment and modify when actual data is available\n",
        "    # df = pd.read_csv(file_path)\n",
        "    # return df\n",
        "    pass\n",
        "\n",
        "def save_model(model, file_path):\n",
        "    \"\"\"Save trained model to file\"\"\"\n",
        "    import joblib\n",
        "    joblib.dump(model, file_path)\n",
        "    print(f\"Model saved to {file_path}\")\n",
        "\n",
        "def load_model(file_path):\n",
        "    \"\"\"Load trained model from file\"\"\"\n",
        "    import joblib\n",
        "    model = joblib.load(file_path)\n",
        "    return model\n",
        "\n",
        "# =============================================================================\n",
        "# Run the Main Pipeline\n",
        "# =============================================================================\n",
        "\n",
        "if __name__ == \"__main__\":\n",
        "    main()"
      ]
    }
  ]
}