{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3169d628",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Step 1: All libraries imported successfully.\n"
     ]
    }
   ],
   "source": [
    "# Importing Libraries\n",
    "\n",
    "# Data Handling\n",
    "import pandas as pd\n",
    "\n",
    "# Data Preprocessing & Splitting\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "# Models\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "# Evaluation Metrics\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report\n",
    "from sklearn.metrics import roc_curve, roc_auc_score, ConfusionMatrixDisplay\n",
    "\n",
    "# Plotting\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Saving Models\n",
    "import joblib\n",
    "\n",
    "# Ignore warnings for cleaner output\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "print(\"All libraries imported successfully.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d1ae5bce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset loaded successfully.\n",
      "First 5 rows of the data:\n",
      "                                               Claim  Label\n",
      "0  Berberine may help improve insulin sensitivity...      1\n",
      "1  Applying a warm compress can ease sinus conges...      1\n",
      "2      Eating garlic supports cardiovascular health.      1\n",
      "3     Peppermint tea may help with mild indigestion.      1\n",
      "4              Echinacea is a proven immune-booster.      0\n"
     ]
    }
   ],
   "source": [
    "# Dataset Loading\n",
    "\n",
    "try:\n",
    "    df = pd.read_csv('dataset.csv', on_bad_lines='skip')\n",
    "    \n",
    "    print(\"Dataset loaded successfully.\")\n",
    "    print(\"First 5 rows of the data:\")\n",
    "    print(df.head())\n",
    "    \n",
    "except FileNotFoundError:\n",
    "    print(\"Error: 'dataset.csv' not found.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "56a57f10",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original data shape: (9007, 2)\n",
      "Unique labels found (before cleaning): [1 0]\n",
      "\n",
      "Unique labels found (after cleaning): [1 0]\n",
      "Missing values after cleaning:\n",
      "Claim    0\n",
      "Label    0\n",
      "dtype: int64\n",
      "Data shape after cleaning: (9007, 2)\n"
     ]
    }
   ],
   "source": [
    "# Data Preprocessing and Cleaning\n",
    "\n",
    "print(f\"Original data shape: {df.shape}\")\n",
    "\n",
    "print(f\"Unique labels found (before cleaning): {df['Label'].unique()}\")\n",
    "\n",
    "df['Label'] = pd.to_numeric(df['Label'], errors='coerce')\n",
    "\n",
    "df.dropna(subset=['Claim', 'Label'], inplace=True)\n",
    "\n",
    "df = df[df['Label'].isin([0, 1])]\n",
    "\n",
    "df['Label'] = df['Label'].astype(int)\n",
    "\n",
    "print(f\"\\nUnique labels found (after cleaning): {df['Label'].unique()}\")\n",
    "print(f\"Missing values after cleaning:\\n{df.isnull().sum()}\")\n",
    "print(f\"Data shape after cleaning: {df.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e49055b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features (X) shape: (9007,)\n",
      "Target (y) shape: (9007,)\n"
     ]
    }
   ],
   "source": [
    "# Features (X) and Target (y)\n",
    "\n",
    "X = df['Claim']  # The raw text data\n",
    "y = df['Label']  # 0 or 1\n",
    "\n",
    "print(f\"Features (X) shape: {X.shape}\")\n",
    "print(f\"Target (y) shape: {y.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4de20c04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set size: 6304 samples\n",
      "Test set size: 2703 samples\n"
     ]
    }
   ],
   "source": [
    "# Spliting Data into Training and Test Sets\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, y, \n",
    "    test_size=0.3,    # 30% for testing\n",
    "    train_size=0.7,   # 70% for training\n",
    "    random_state=42,  # For reproducible results\n",
    "    shuffle=True\n",
    ")\n",
    "\n",
    "print(f\"Training set size: {X_train.shape[0]} samples\")\n",
    "print(f\"Test set size: {X_test.shape[0]} samples\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "8a493bbf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Text vectorization complete.\n",
      "Shape of vectorized training data: (6304, 2789)\n",
      "Shape of vectorized test data: (2703, 2789)\n"
     ]
    }
   ],
   "source": [
    "# Vectorize Text & Clean Symbols (TF-IDF)\n",
    "\n",
    "# Initialize the vectorizer. \n",
    "vectorizer = TfidfVectorizer(stop_words='english', max_features=5000)\n",
    "\n",
    "# Fit and transform the training data\n",
    "X_train_vec = vectorizer.fit_transform(X_train)\n",
    "\n",
    "# Transform the test data\n",
    "X_test_vec = vectorizer.transform(X_test)\n",
    "\n",
    "print(\"Text vectorization complete.\")\n",
    "print(f\"Shape of vectorized training data: {X_train_vec.shape}\")\n",
    "print(f\"Shape of vectorized test data: {X_test_vec.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "cf6fd882",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Logistic Regression model...\n",
      "Logistic Regression training complete.\n"
     ]
    }
   ],
   "source": [
    "# Training & Prediction using Logistic Regression\n",
    "\n",
    "print(\"Training Logistic Regression model...\")\n",
    "model_lr = LogisticRegression(random_state=42)\n",
    "\n",
    "# Model training on the vectorized training data\n",
    "model_lr.fit(X_train_vec, y_train)\n",
    "\n",
    "# Get binary predictions (0 or 1) for the test set\n",
    "y_pred_lr = model_lr.predict(X_test_vec)\n",
    "\n",
    "# Probability predictions for the AUC-ROC curve\n",
    "y_proba_lr = model_lr.predict_proba(X_test_vec)[:, 1] \n",
    "\n",
    "print(\"Logistic Regression training complete.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "70b19146",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Logistic Regression Evaluation ---\n",
      "Accuracy: 0.9523\n",
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Fake (0)       0.97      0.94      0.95      1348\n",
      "    Real (1)       0.94      0.97      0.95      1355\n",
      "\n",
      "    accuracy                           0.95      2703\n",
      "   macro avg       0.95      0.95      0.95      2703\n",
      "weighted avg       0.95      0.95      0.95      2703\n",
      "\n",
      "\n",
      "Confusion Matrix:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Evaluating Logistic Regression Metrics\n",
    "\n",
    "print(\"--- Logistic Regression Evaluation ---\")\n",
    "\n",
    "# Accuracy\n",
    "print(f\"Accuracy: {accuracy_score(y_test, y_pred_lr):.4f}\")\n",
    "\n",
    "# Classification Report\n",
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_test, y_pred_lr, labels=[0, 1], target_names=['Fake (0)', 'Real (1)']))\n",
    "\n",
    "# Confusion Matrix\n",
    "print(\"\\nConfusion Matrix:\")\n",
    "cm_lr = confusion_matrix(y_test, y_pred_lr, labels=[0, 1])\n",
    "\n",
    "# Plot the matrix\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm_lr, \n",
    "                              display_labels=['Fake (0)', 'Real (1)'])\n",
    "disp.plot(cmap=plt.cm.Blues, values_format='d')\n",
    "plt.title(\"Logistic Regression Confusion Matrix\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "64032d01",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression AUC Score: 0.9907\n"
     ]
    }
   ],
   "source": [
    "# Logistic Regression AUC-ROC\n",
    "\n",
    "# AUC score\n",
    "auc_lr = roc_auc_score(y_test, y_proba_lr)\n",
    "print(f\"Logistic Regression AUC Score: {auc_lr:.4f}\")\n",
    "\n",
    "# ROC curve\n",
    "fpr_lr, tpr_lr, _ = roc_curve(y_test, y_proba_lr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "76949c64",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training SVM model...\n",
      "SVM training complete.\n"
     ]
    }
   ],
   "source": [
    "# Training & Prediction using SVM\n",
    "\n",
    "print(\"Training SVM model...\")\n",
    "\n",
    "# Initializing\n",
    "model_svm = SVC(random_state=42, probability=True)\n",
    "\n",
    "# Model trainingodel\n",
    "model_svm.fit(X_train_vec, y_train)\n",
    "\n",
    "# Binary predictions (0 or 1)\n",
    "y_pred_svm = model_svm.predict(X_test_vec)\n",
    "\n",
    "# Probability predictions (AUC-ROC curve)\n",
    "y_proba_svm = model_svm.predict_proba(X_test_vec)[:, 1]\n",
    "\n",
    "print(\"SVM training complete.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "328f961e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM Evaluation\n",
      "Accuracy: 0.9612\n",
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Fake (0)       0.97      0.95      0.96      1348\n",
      "    Real (1)       0.95      0.97      0.96      1355\n",
      "\n",
      "    accuracy                           0.96      2703\n",
      "   macro avg       0.96      0.96      0.96      2703\n",
      "weighted avg       0.96      0.96      0.96      2703\n",
      "\n",
      "\n",
      "Confusion Matrix:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Evaluateing SVM Metrics\n",
    "\n",
    "print(\"SVM Evaluation\")\n",
    "\n",
    "# Accuracy\n",
    "print(f\"Accuracy: {accuracy_score(y_test, y_pred_svm):.4f}\")\n",
    "\n",
    "# Classification Report\n",
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_test, y_pred_svm, labels=[0, 1], target_names=['Fake (0)', 'Real (1)']))\n",
    "\n",
    "# Graphical Confusion Matrix\n",
    "print(\"\\nConfusion Matrix:\")\n",
    "cm_svm = confusion_matrix(y_test, y_pred_svm, labels=[0, 1])\n",
    "\n",
    "# Matrix ploting\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm_svm, \n",
    "                              display_labels=['Fake (0)', 'Real (1)'])\n",
    "disp.plot(cmap=plt.cm.Blues, values_format='d')\n",
    "plt.title(\"SVM Confusion Matrix\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "306e0617",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM AUC Score: 0.9938\n"
     ]
    }
   ],
   "source": [
    "# Calculating SVM AUC-ROC\n",
    "\n",
    "# AUC score\n",
    "auc_svm = roc_auc_score(y_test, y_proba_svm)\n",
    "print(f\"SVM AUC Score: {auc_svm:.4f}\")\n",
    "\n",
    "# ROC curve\n",
    "fpr_svm, tpr_svm, _ = roc_curve(y_test, y_proba_svm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "be05bc1d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Random Forest model...\n",
      "Random Forest training complete.\n"
     ]
    }
   ],
   "source": [
    "# Training & Prediction using Random Forest\n",
    "\n",
    "print(\"Training Random Forest model...\")\n",
    "\n",
    "# Initialize the model\n",
    "model_rf = RandomForestClassifier(random_state=42, n_estimators=100)\n",
    "\n",
    "# Train the model\n",
    "model_rf.fit(X_train_vec, y_train)\n",
    "\n",
    "# Binary predictions (0 or 1)\n",
    "y_pred_rf = model_rf.predict(X_test_vec)\n",
    "\n",
    "# Probability predictions for the AUC-ROC curve\n",
    "y_proba_rf = model_rf.predict_proba(X_test_vec)[:, 1]\n",
    "\n",
    "print(\"Random Forest training complete.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "65136e05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Evaluation\n",
      "Accuracy: 0.9482\n",
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Fake (0)       0.95      0.94      0.95      1348\n",
      "    Real (1)       0.94      0.95      0.95      1355\n",
      "\n",
      "    accuracy                           0.95      2703\n",
      "   macro avg       0.95      0.95      0.95      2703\n",
      "weighted avg       0.95      0.95      0.95      2703\n",
      "\n",
      "\n",
      "Confusion Matrix:\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Evaluating Random Forest Metrics\n",
    "\n",
    "print(\"Random Forest Evaluation\")\n",
    "\n",
    "# Accuracy\n",
    "print(f\"Accuracy: {accuracy_score(y_test, y_pred_rf):.4f}\")\n",
    "\n",
    "# Classification Report\n",
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_test, y_pred_rf, labels=[0, 1], target_names=['Fake (0)', 'Real (1)']))\n",
    "\n",
    "# Graphical Confusion Matrix\n",
    "print(\"\\nConfusion Matrix:\")\n",
    "cm_rf = confusion_matrix(y_test, y_pred_rf, labels=[0, 1])\n",
    "\n",
    "# Plot the matrix\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm_rf, \n",
    "                              display_labels=['Fake (0)', 'Real (1)'])\n",
    "disp.plot(cmap=plt.cm.Blues, values_format='d')\n",
    "plt.title(\"Random Forest Confusion Matrix\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "26eae5fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest AUC Score: 0.9905\n"
     ]
    }
   ],
   "source": [
    "# Calculating Random Forest AUC-ROC\n",
    "\n",
    "# AUC score\n",
    "auc_rf = roc_auc_score(y_test, y_proba_rf)\n",
    "print(f\"Random Forest AUC Score: {auc_rf:.4f}\")\n",
    "\n",
    "# ROC curve\n",
    "fpr_rf, tpr_rf, _ = roc_curve(y_test, y_proba_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "4a1440cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Comparison Summary\n",
      "                     Accuracy  AUC Score  F1-Score (Real)\n",
      "Model                                                    \n",
      "SVM                  0.961154   0.993813         0.961553\n",
      "Logistic Regression  0.952275   0.990729         0.953074\n",
      "Random Forest        0.948206   0.990516         0.948567\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Model Comparison\n",
    "\n",
    "# 1. Classification reports\n",
    "report_lr = classification_report(y_test, y_pred_lr, labels=[0, 1], output_dict=True)\n",
    "report_svm = classification_report(y_test, y_pred_svm, labels=[0, 1], output_dict=True)\n",
    "report_rf = classification_report(y_test, y_pred_rf, labels=[0, 1], output_dict=True)\n",
    "\n",
    "# 2. F1-scores\n",
    "f1_lr = report_lr['1']['f1-score']\n",
    "f1_svm = report_svm['1']['f1-score']\n",
    "f1_rf = report_rf['1']['f1-score']\n",
    "\n",
    "# Comparing DataFrame using the accuracy, auc, and f1 variables\n",
    "comparison_data = {\n",
    "    'Model': ['Logistic Regression', 'SVM', 'Random Forest'],\n",
    "    'Accuracy': [accuracy_score(y_test, y_pred_lr), accuracy_score(y_test, y_pred_svm), accuracy_score(y_test, y_pred_rf)],\n",
    "    'AUC Score': [auc_lr, auc_svm, auc_rf],\n",
    "    'F1-Score (Real)': [f1_lr, f1_svm, f1_rf]\n",
    "}\n",
    "\n",
    "comparison_df = pd.DataFrame(comparison_data)\n",
    "comparison_df = comparison_df.set_index('Model')\n",
    "\n",
    "print(\"Model Comparison Summary\")\n",
    "# Sort by AUC Score, from best to worst\n",
    "print(comparison_df.sort_values(by='AUC Score', ascending=False))\n",
    "\n",
    "# Ploting ROC Curves for All Models\n",
    "plt.figure(figsize=(10, 8))\n",
    "plt.plot(fpr_lr, tpr_lr, label=f'Logistic Regression (AUC = {auc_lr:.3f})')\n",
    "plt.plot(fpr_svm, tpr_svm, label=f'SVM (AUC = {auc_svm:.3f})')\n",
    "plt.plot(fpr_rf, tpr_rf, label=f'Random Forest (AUC = {auc_rf:.3f})')\n",
    "plt.plot([0, 1], [0, 1], 'r--', label='Random Guess (AUC = 0.5)') \n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Model Comparison: ROC Curves')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "3b6991fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best model ('SVC') saved to best_model.joblib\n",
      "Vectorizer saved to vectorizer.joblib\n"
     ]
    }
   ],
   "source": [
    "# Saving the Best Model & Vectorizer\n",
    "\n",
    "#!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n",
    "# Choose your best model here\n",
    "# Exp: best_model = model_rf, model_svm, model_lr\n",
    "best_model = model_svm \n",
    "\n",
    "model_filename = 'best_model.joblib'\n",
    "vectorizer_filename = 'vectorizer.joblib'\n",
    "\n",
    "# Save the model\n",
    "joblib.dump(best_model, model_filename)\n",
    "\n",
    "# Save the vectorizer\n",
    "joblib.dump(vectorizer, vectorizer_filename)\n",
    "\n",
    "print(f\"Best model ('{best_model.__class__.__name__}') saved to {model_filename}\")\n",
    "print(f\"Vectorizer saved to {vectorizer_filename}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "839e1b3b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VERA - Verifying Remedies Assertion\n",
      "Model and vectorizer loaded successfully.\n",
      "Enter a REMEDY to check its assertion.\n",
      "\n",
      "Statement: 'turmeric heals wounds'\n",
      "Prediction: Fake (Misinformation) (Confidence: 98.91%)\n",
      "----------------------------------------\n",
      "\n",
      "Statement: 'fennel seeds helps in bloating'\n",
      "Prediction: Real (Evidence-based) (Confidence: 96.65%)\n",
      "----------------------------------------\n",
      "\n",
      "Statement: 'turmeric cure wounds'\n",
      "Prediction: Fake (Misinformation) (Confidence: 99.95%)\n",
      "----------------------------------------\n",
      "Exiting detector. Goodbye!\n"
     ]
    }
   ],
   "source": [
    "# Interactive Prediction\n",
    "\n",
    "try:\n",
    "    # 1. Loading saved model and vectorizer\n",
    "    loaded_model = joblib.load('best_model.joblib')\n",
    "    loaded_vectorizer = joblib.load('vectorizer.joblib')\n",
    "    print(\"VERA - Verifying Remedies Assertion\")\n",
    "    print(\"Model and vectorizer loaded successfully.\")\n",
    "    print(\"Enter a REMEDY to check its assertion.\")\n",
    "    \n",
    "    # 2. Create an infinite loop to keep asking for input\n",
    "    while True:\n",
    "        # 3. User input\n",
    "        user_statement = input(\"\\nEnter a claim to check (or type 'quit' to exit): \")\n",
    "        \n",
    "        # 4. Quit check\n",
    "        if user_statement.lower() == 'quit':\n",
    "            print(\"Exiting detector. Goodbye!\")\n",
    "            break\n",
    "            \n",
    "        # 5. Transform the single statement\n",
    "        new_statement_list = [user_statement]\n",
    "        new_statement_vec = loaded_vectorizer.transform(new_statement_list)\n",
    "        \n",
    "        # 6. Predict using the loaded model\n",
    "        prediction = loaded_model.predict(new_statement_vec)[0]\n",
    "        \n",
    "        probabilities = loaded_model.predict_proba(new_statement_vec)[0]\n",
    "        \n",
    "        # 7. Print result\n",
    "        if prediction == 1:\n",
    "            result = \"Real (Evidence-based)\"\n",
    "            confidence = probabilities[1] * 100 # Probability of class 1\n",
    "        else:\n",
    "            result = \"Fake (Misinformation)\"\n",
    "            confidence = probabilities[0] * 100 # Probability of class 0\n",
    "            \n",
    "        print(f\"\\nStatement: '{user_statement}'\") \n",
    "\n",
    "        print(f\"Prediction: {result} (Confidence: {confidence:.2f}%)\")\n",
    "        print(\"-\" * 40)\n",
    "\n",
    "except FileNotFoundError:\n",
    "    print(\"\\nError: Could not load saved model files ('best_model.joblib' or 'vectorizer.joblib').\")\n",
    "    print(\"Please ensure Step 17 was run successfully before running this step.\")\n",
    "except Exception as e:\n",
    "    print(f\"An error occurred: {e}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "daf056db",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}