customer_churn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Customer Churn Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing packages\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
    "from sklearn.metrics import roc_curve, roc_auc_score, auc\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Defining Churn for analysis\n",
    "Churn: (no interaction in the last 20 days) through T/F -> 1/0\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "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>user_id</th>\n",
       "      <th>song_id</th>\n",
       "      <th>play_count</th>\n",
       "      <th>last_played</th>\n",
       "      <th>user_age</th>\n",
       "      <th>user_country</th>\n",
       "      <th>days_since_last_played</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>00Srvq2BtgG9afXY9kzB1V</td>\n",
       "      <td>27</td>\n",
       "      <td>2024-03-29 11:25:10.912694</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>01N3UDXUKUP6e2It26WAak</td>\n",
       "      <td>5</td>\n",
       "      <td>2024-05-14 11:25:10.912733</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>09CnYHiZ5jGT1wr1TXJ9Zt</td>\n",
       "      <td>6</td>\n",
       "      <td>2024-05-29 11:25:10.912735</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0ClPIeT6MSgfSgQ9ZrJbAq</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-04-27 11:25:10.912737</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0CnlKa2jjIMzG0gLN1KZbD</td>\n",
       "      <td>14</td>\n",
       "      <td>2024-03-12 11:25:10.912738</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>83</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0HGItm7Ox9AnHZbx26nkE4</td>\n",
       "      <td>12</td>\n",
       "      <td>2024-05-31 11:25:10.912739</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0Jx0wecBc4647J4QMdmsY1</td>\n",
       "      <td>4</td>\n",
       "      <td>2024-03-25 11:25:10.912740</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>70</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0LMwmV37RCmBO2so0szAFs</td>\n",
       "      <td>8</td>\n",
       "      <td>2024-03-29 11:25:10.912741</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0M0ANKNzmM4Odd7FNKghzW</td>\n",
       "      <td>48</td>\n",
       "      <td>2024-04-09 11:25:10.912742</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>55</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>india_user_0</td>\n",
       "      <td>0TL0LFcwIBF5eX7arDIKxY</td>\n",
       "      <td>12</td>\n",
       "      <td>2024-03-18 11:25:10.912743</td>\n",
       "      <td>20</td>\n",
       "      <td>india</td>\n",
       "      <td>77</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        user_id                 song_id  play_count  \\\n",
       "0  india_user_0  00Srvq2BtgG9afXY9kzB1V          27   \n",
       "1  india_user_0  01N3UDXUKUP6e2It26WAak           5   \n",
       "2  india_user_0  09CnYHiZ5jGT1wr1TXJ9Zt           6   \n",
       "3  india_user_0  0ClPIeT6MSgfSgQ9ZrJbAq           1   \n",
       "4  india_user_0  0CnlKa2jjIMzG0gLN1KZbD          14   \n",
       "5  india_user_0  0HGItm7Ox9AnHZbx26nkE4          12   \n",
       "6  india_user_0  0Jx0wecBc4647J4QMdmsY1           4   \n",
       "7  india_user_0  0LMwmV37RCmBO2so0szAFs           8   \n",
       "8  india_user_0  0M0ANKNzmM4Odd7FNKghzW          48   \n",
       "9  india_user_0  0TL0LFcwIBF5eX7arDIKxY          12   \n",
       "\n",
       "                 last_played  user_age user_country  days_since_last_played  \\\n",
       "0 2024-03-29 11:25:10.912694        20        india                      66   \n",
       "1 2024-05-14 11:25:10.912733        20        india                      20   \n",
       "2 2024-05-29 11:25:10.912735        20        india                       5   \n",
       "3 2024-04-27 11:25:10.912737        20        india                      37   \n",
       "4 2024-03-12 11:25:10.912738        20        india                      83   \n",
       "5 2024-05-31 11:25:10.912739        20        india                       3   \n",
       "6 2024-03-25 11:25:10.912740        20        india                      70   \n",
       "7 2024-03-29 11:25:10.912741        20        india                      66   \n",
       "8 2024-04-09 11:25:10.912742        20        india                      55   \n",
       "9 2024-03-18 11:25:10.912743        20        india                      77   \n",
       "\n",
       "   churn  \n",
       "0      0  \n",
       "1      0  \n",
       "2      0  \n",
       "3      0  \n",
       "4      0  \n",
       "5      0  \n",
       "6      0  \n",
       "7      0  \n",
       "8      0  \n",
       "9      0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the dataset\n",
    "data = pd.read_csv('data/user/user_data.csv')\n",
    "\n",
    "# convert 'last_played' to datetime and create days since last played\n",
    "data['last_played'] = pd.to_datetime(data['last_played'])\n",
    "data['days_since_last_played'] = (pd.to_datetime('now') - data['last_played']).dt.days\n",
    "\n",
    "# determine churn, per user NOT per row\n",
    "user_last_played = data.groupby('user_id')['days_since_last_played'].max().reset_index() # Create a df showing closest streaming date per user\n",
    "user_last_played['churn'] = user_last_played['days_since_last_played'] > 20 # Define churn\n",
    "\n",
    "# Add randomness to the churn, adding variability to mimic reality\n",
    "np.random.seed(42)  # reproducibility\n",
    "flip_prob = 0.20  # Probability churn changing\n",
    "random_flips = np.random.rand(len(user_last_played)) < flip_prob\n",
    "user_last_played['churn'] = user_last_played['churn'] ^ random_flips\n",
    "\n",
    "# merge the churn information back into the original dataset\n",
    "data = data.merge(user_last_played[['user_id', 'churn']], on='user_id', how='left')\n",
    "\n",
    "# convert churn from T/F to 1/0\n",
    "data['churn'] = data['churn'].astype(int)\n",
    "\n",
    "data.to_csv(\"data/user/user_churn.csv\", index=False)\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "churn\n",
      "1    775\n",
      "0    225\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Group by 'user_id' and take the maximum 'churn' value per user\n",
    "user_churn = data.groupby('user_id')['churn'].max().reset_index()\n",
    "\n",
    "# Count the number of users with churn = 0 and churn = 1\n",
    "churn_counts = user_churn['churn'].value_counts()\n",
    "\n",
    "print(churn_counts)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define features and target\n",
    "features = ['user_age', 'play_count']\n",
    "target = 'churn'\n",
    "\n",
    "X = data[features]\n",
    "y = data[target]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the data\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters found: {'bootstrap': True, 'max_depth': 10, 'min_samples_leaf': 6, 'min_samples_split': 10, 'n_estimators': 100}\n"
     ]
    }
   ],
   "source": [
    "# Hyperparameter Tuning using GridSearchCV\n",
    "param_grid = {\n",
    "    'n_estimators': [50,100],\n",
    "    'max_depth': [5,10],\n",
    "    'min_samples_split': [10,15],\n",
    "    'min_samples_leaf': [4,6],\n",
    "    'bootstrap': [True]\n",
    "}\n",
    "\n",
    "rf = RandomForestClassifier(random_state=42)\n",
    "grid_search = GridSearchCV(estimator=rf, param_grid=param_grid, cv=3, n_jobs=-1, verbose=0, scoring='roc_auc')\n",
    "grid_search.fit(X_train, y_train)\n",
    "\n",
    "# Best model from GridSearch\n",
    "best_rf = grid_search.best_estimator_\n",
    "print(f'Best parameters found: {grid_search.best_params_}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7733\n",
      "Precision: 0.77349\n",
      "Recall: 0.99966\n",
      "F1 Score: 0.87215\n",
      "Confusion Matrix: \n",
      " [[    4  9571]\n",
      " [   11 32684]]\n"
     ]
    }
   ],
   "source": [
    "# Predict and evaluate\n",
    "y_pred = best_rf.predict(X_test)\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "precision = precision_score(y_test, y_pred)\n",
    "recall = recall_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred)\n",
    "\n",
    "print(f'Accuracy: {accuracy:.4f}')\n",
    "print(f'Precision: {precision:.5f}')\n",
    "print(f'Recall: {recall:.5f}')\n",
    "print(f'F1 Score: {f1:.5f}')\n",
    "print(f'Confusion Matrix: \\n {confusion_matrix(y_test, y_pred)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualising the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Make predictions\n",
    "y_pred_proba = best_rf.predict_proba(X_test)[:, 1]\n",
    "\n",
    "# Calculate ROC curve and AUC\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_proba)\n",
    "roc_auc = auc(fpr, tpr)\n",
    "\n",
    "# Plot the ROC curve\n",
    "plt.figure(figsize=(10, 6)) # making the figure larger\n",
    "plt.plot(fpr, tpr, color='darkblue', lw=2, label=f'ROC AUC: {roc_auc:.3f}')\n",
    "plt.plot([0, 1], [0, 1], color='red', lw=2, linestyle='--') # diagonal line for ref\n",
    "plt.xlim([0.0, 1.0]) # correct limits\n",
    "plt.ylim([0.0, 1.05]) # limits + extra for visual aid\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.savefig('figures/churn_roc')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/churn_prediction_model.pkl']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Save the model\n",
    "import joblib\n",
    "joblib.dump(best_rf, 'models/churn_prediction_model.pkl')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.12.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}