Supervised Learning - Gaussian NB and Random Forest.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>In this document, we will follow a supervised learning approach for classifying the events to normal and anomalies.</h1> (Outlier is a synonym for Anomaly in Data Science language)\n",
    "\n",
    "<p>The dataset has 8 features consisting of few outliers in it.</p>\n",
    "\n",
    "<b>What are we looking for?</b> : In anomaly detection, the Recall is very important, even more important than Accuracy. This is because, the percentage of an outlier in any dataset is generally very small. And that \"very small\" amount of anomaly should be detected correctly as well, which is determined by Recall rate. You will see in the models below.\n",
    "\n",
    "To end this dosument, we will compare the results of all the three approaches on the basis of their Recall and Accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Author: Kopal Soni<br>\n",
    "E-mail: soni.kopal@gmail.com<br>\n",
    "LinkedIn: https://www.linkedin.com/in/kopal-soni-72992810b/<br>\n",
    "Github: https://github.com/ksoni5<br>\n",
    "Importing the required libraries<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix, recall_score, f1_score, precision_score, accuracy_score\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Datasets\n",
    "\"\"\"\n",
    "data = pd.read_csv('Dataset/experiment_contains_anomalies.csv')\n",
    "\n",
    "\"\"\"Random Split\"\"\"\n",
    "X = data.drop(['Sds_Armed', 'Anomaly_Tag'], axis=1)\n",
    "Y = pd.DataFrame(data['Anomaly_Tag'], columns=['Anomaly_Tag'])\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, Y, train_size=0.7, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.7957</td>\n",
       "      <td>6.1964</td>\n",
       "      <td>508.57</td>\n",
       "      <td>875.40</td>\n",
       "      <td>0.43922</td>\n",
       "      <td>1.5473</td>\n",
       "      <td>57.021</td>\n",
       "      <td>56.914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.7851</td>\n",
       "      <td>6.1508</td>\n",
       "      <td>503.90</td>\n",
       "      <td>867.18</td>\n",
       "      <td>0.43947</td>\n",
       "      <td>1.3525</td>\n",
       "      <td>53.658</td>\n",
       "      <td>58.856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1.7760</td>\n",
       "      <td>6.1173</td>\n",
       "      <td>499.62</td>\n",
       "      <td>860.56</td>\n",
       "      <td>0.42528</td>\n",
       "      <td>1.3391</td>\n",
       "      <td>52.551</td>\n",
       "      <td>69.101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1.7641</td>\n",
       "      <td>6.0799</td>\n",
       "      <td>494.07</td>\n",
       "      <td>851.51</td>\n",
       "      <td>0.42802</td>\n",
       "      <td>1.3679</td>\n",
       "      <td>51.228</td>\n",
       "      <td>75.935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1.7525</td>\n",
       "      <td>6.0333</td>\n",
       "      <td>486.73</td>\n",
       "      <td>840.86</td>\n",
       "      <td>0.40701</td>\n",
       "      <td>1.3128</td>\n",
       "      <td>47.228</td>\n",
       "      <td>72.784</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       X1      X2      X3      X4       X5      X6      X7      X8\n",
       "0  1.7957  6.1964  508.57  875.40  0.43922  1.5473  57.021  56.914\n",
       "1  1.7851  6.1508  503.90  867.18  0.43947  1.3525  53.658  58.856\n",
       "2  1.7760  6.1173  499.62  860.56  0.42528  1.3391  52.551  69.101\n",
       "3  1.7641  6.0799  494.07  851.51  0.42802  1.3679  51.228  75.935\n",
       "4  1.7525  6.0333  486.73  840.86  0.40701  1.3128  47.228  72.784"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Anomaly_Tag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anomaly_Tag\n",
       "0            0\n",
       "1            0\n",
       "2            0\n",
       "3            0\n",
       "4            0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>Let's start with Naive Bayes which belongs to linear model class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score: 84.819% Recall Score: 33.247% Precision : 45.168% and F1 Score: 38.302%\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/sklearn/utils/validation.py:73: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(**kwargs)\n"
     ]
    }
   ],
   "source": [
    "gaussianNB = GaussianNB()\n",
    "gaussianNB.fit(x_train, y_train)\n",
    "gnbPredict = gaussianNB.predict(x_test)\n",
    "\n",
    "print(\"Accuracy Score: {:.3f}% Recall Score: {:.3f}% Precision : {:.3f}% and F1 Score: {:.3f}%\"\n",
    "      .format(accuracy_score(y_test, gnbPredict)*100, recall_score(y_test, gnbPredict)*100, \n",
    "              precision_score(y_test, gnbPredict)*100, f1_score(y_test, gnbPredict)*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(confusion_matrix(y_test, gnbPredict), annot=True, cmap='jet')\n",
    "plt.xlabel(\"Actual\")\n",
    "plt.ylabel(\"Predicted\")\n",
    "plt.title(\"Confusion Matrix\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20,5))\n",
    "plt.scatter(x_test.iloc[:,0], y_test, color=\"blue\", marker='o', s=400 , alpha=0.1, label='Actual')\n",
    "plt.scatter(x_test.iloc[:,0], gnbPredict, color=\"red\", marker='.', s=100, alpha=0.4, label='Predicted')\n",
    "plt.title(\"Actual (blue) vs Predicted (red) on the most important feature X1 using Gaussian Naive Bayes\")\n",
    "plt.xlabel(\"X1 - Most important Feature\")\n",
    "plt.ylabel(\"Actual\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can observe in the above plot that some of the red dots for outlier (1) are wrongly predicted.\n",
    "\n",
    "Recall from Gaussian NB model (~ 33%) wasn't acceptable at all. From the confusion matrix, we can see that a lot of normal data points were corretly predicted but in 'Anomaly detection', correct prediction of an anomaly is as important as normal data point. And so, the ~85% accuracy doesn't mean that the model is accepatble. We need to look for a model which can Recall both, normal as well as outliers, more accurately."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>Let's check out the champs, Random Forest which belongs to ensemble class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/ipykernel_launcher.py:6: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score: 98.568% Recall Score: 91.082% Precision : 98.715% and F1 Score: 94.745%\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Random Forest Classifier\n",
    "\"\"\"\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators = 10, random_state=42)\n",
    "rf.fit(x_train, y_train)\n",
    "rfPredict = rf.predict(x_test)\n",
    "\n",
    "print(\"Accuracy Score: {:.3f}% Recall Score: {:.3f}% Precision : {:.3f}% and F1 Score: {:.3f}%\"\n",
    "      .format(accuracy_score(y_test, rfPredict)*100, recall_score(y_test, rfPredict)*100, \n",
    "              precision_score(y_test, rfPredict)*100, f1_score(y_test, rfPredict)*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_importance = pd.Series(rf.feature_importances_, index = x_train.columns)\n",
    "feature_importance.plot(kind='barh', color=['red', 'blue', 'black', 'magenta', 'cyan', 'green', 'yellow', 'crimson'])\n",
    "plt.title('Feature Importance from Random Forest')\n",
    "plt.ylabel(\"Features\")\n",
    "plt.xlabel(\"Importance\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(confusion_matrix(y_test, rfPredict), annot=True, cmap=\"jet\")\n",
    "plt.xlabel(\"Actual\")\n",
    "plt.ylabel(\"Predicted\")\n",
    "plt.title(\"Confusion Matrix\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "The scatter plot shows the actual vs predicted data points for feature 'X1'\n",
    "\"\"\"\n",
    "fig = plt.figure(figsize=(20,5))\n",
    "plt.scatter(x_test.iloc[:,0], y_test, color=\"blue\", marker='o', s=400 , alpha=0.1, label='Actual')\n",
    "plt.scatter(x_test.iloc[:,0], rfPredict, color=\"red\", marker='.', s=100, alpha=0.4, label='Predicted')\n",
    "plt.title(\"Actual (blue) vs Predicted (red) on the most important feature X1 using Random Forest Classifier\")\n",
    "plt.xlabel(\"X1 - Most important Feature\")\n",
    "plt.ylabel(\"Actual\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h4> Woah! Impressive, aren't they. This model shows that very few outliers and normal data points were incorrectly detected."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<H3>Well we do not have much to say. The results are clearly evident as the Random Forests gave us amazing results."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.0 64-bit",
   "language": "python",
   "name": "python37064bitc68a0a7a3ca348988f11c9fc9f736202"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}