Naive Bayes Classifier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Naive Bayes Classifier\n",
    "\n",
    "By Rama Vempati\n",
    "\n",
    "To connect with or follow me on LinkedIn [Click Here](https://www.linkedin.com/in/ramavempati/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "\n",
    "* Naive Bayes models are a group of extremely fast and simple classification algorithms that are often suitable for very high-dimensional datasets.\n",
    "\n",
    "\n",
    "* Because they are so fast and have so few tunable parameters, they end up being very useful as a quick-and-dirty baseline for a classification problem."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bayesian Classification\n",
    "\n",
    "Naive Bayes classifiers are built on Bayesian classification methods.\n",
    "These rely on Bayes's theorem, which is an equation describing the relationship of conditional probabilities of statistical quantities.\n",
    "In Bayesian classification, we're interested in finding the probability of a label given some observed features, which we can write as $P(L~|~{\\rm features})$.\n",
    "Bayes's theorem tells us how to express this in terms of quantities we can compute more directly:\n",
    "\n",
    "$$\n",
    "P(L~|~{\\rm features}) = \\frac{P({\\rm features}~|~L)P(L)}{P({\\rm features})}\n",
    "$$\n",
    "\n",
    "\n",
    "If we are trying to decide between two labels—let's call them $L_1$ and $L_2$—then one way to make this decision is to compute the ratio of the posterior probabilities for each label:\n",
    "\n",
    "$$\n",
    "\\frac{P(L_1~|~{\\rm features})}{P(L_2~|~{\\rm features})} = \\frac{P({\\rm features}~|~L_1)}{P({\\rm features}~|~L_2)}\\frac{P(L_1)}{P(L_2)}\n",
    "$$\n",
    "\n",
    "\n",
    "All we need now is some model by which we can compute $P({\\rm features}~|~L_i)$ for each label.\n",
    "Such a model is called a *generative model* because it specifies the hypothetical random process that generates the data.\n",
    "Specifying this generative model for each label is the main piece of the training of such a Bayesian classifier.\n",
    "The general version of such a training step is a very difficult task, but we can make it simpler through the use of some simplifying assumptions about the form of this model.\n",
    "\n",
    "This is where the \"naive\" in \"naive Bayes\" comes in: if we make very naive assumptions about the generative model for each label, we can find a rough approximation of the generative model for each class, and then proceed with the Bayesian classification."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Example 1 - Naive Bayes Classification on MNIST dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns; \n",
    "\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ramakrishnavempati/anaconda3/lib/python3.7/site-packages/sklearn/utils/deprecation.py:77: DeprecationWarning: Function fetch_mldata is deprecated; fetch_mldata was deprecated in version 0.20 and will be removed in version 0.22\n",
      "  warnings.warn(msg, category=DeprecationWarning)\n",
      "/Users/ramakrishnavempati/anaconda3/lib/python3.7/site-packages/sklearn/utils/deprecation.py:77: DeprecationWarning: Function mldata_filename is deprecated; mldata_filename was deprecated in version 0.20 and will be removed in version 0.22\n",
      "  warnings.warn(msg, category=DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "# Fetch the MNIST handwritten digit dataset\n",
    "from sklearn.datasets import fetch_mldata\n",
    "mnist = fetch_mldata('MNIST original', data_home=\"data\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000, 784)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Display the number of samples \n",
    "mnist.data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%pylab inline\n",
    "\n",
    "# Display 9 number randomly selectly\n",
    "for c in range(1, 10):\n",
    "    subplot(3, 3,c)\n",
    "    i = randint(mnist.data.shape[0])\n",
    "    im = mnist.data[i].reshape((28,28))\n",
    "    axis(\"off\")\n",
    "    title(\"Label = {}\".format(mnist.target[i]))\n",
    "    imshow(im, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the data into training and test data\n",
    "from sklearn.model_selection import train_test_split\n",
    "x_train, x_test, y_train, y_test = train_test_split(mnist.data, mnist.target, test_size=0.05, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the Multinomial Naive Bayes Classifier\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "clf = MultinomialNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy = \n",
      "81.71428571428572\n"
     ]
    }
   ],
   "source": [
    "# Perform the predictions\n",
    "clf.fit(x_train, y_train)\n",
    "\n",
    "# Perform the predictions\n",
    "y_predicted = clf.predict(x_test)\n",
    "\n",
    "# Calculate the accuracy of the prediction\n",
    "from sklearn.metrics import accuracy_score\n",
    "print(\"Accuracy = \")\n",
    "print(format(accuracy_score(y_test, y_predicted)*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.91      0.89      0.90       320\n",
      "           1       0.85      0.95      0.90       387\n",
      "           2       0.89      0.84      0.87       352\n",
      "           3       0.78      0.81      0.80       356\n",
      "           4       0.83      0.73      0.78       329\n",
      "           5       0.85      0.67      0.75       342\n",
      "           6       0.90      0.91      0.91       377\n",
      "           7       0.94      0.82      0.87       360\n",
      "           8       0.63      0.72      0.67       337\n",
      "           9       0.66      0.80      0.72       340\n",
      "\n",
      "   micro avg       0.82      0.82      0.82      3500\n",
      "   macro avg       0.82      0.81      0.82      3500\n",
      "weighted avg       0.83      0.82      0.82      3500\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Cross validate the scores\n",
    "from sklearn.metrics import classification_report\n",
    "print(\"Classification Report\")\n",
    "print(format(classification_report(y_test, y_predicted, labels=range(0,10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Understanding of above metrics\n",
    "\n",
    "\n",
    "True positive and true negatives are the observations that are correctly predicted and therefore shown in green. We want to minimize false positives and false negatives so they are shown in red color.\n",
    "\n",
    "<img src='images/accuracy-table.png' alt='accuracy-table' style=\"width: 500px;\"/>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**True Positives (TP)** - These are the correctly predicted positive values which means that the value of actual class is yes and the value of predicted class is also yes.\n",
    "\n",
    "**True Negatives (TN)** - These are the correctly predicted negative values which means that the value of actual class is no and value of predicted class is also no.\n",
    "\n",
    "False positives and false negatives, these values occur when your actual class contradicts with the predicted class.\n",
    "\n",
    "**False Positives (FP)** – When actual class is no and predicted class is yes.\n",
    "\n",
    "**False Negatives (FN)** – When actual class is yes but predicted class in no."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Accuracy** - Accuracy is the most intuitive performance measure and it is simply a ratio of correctly predicted observation to the total observations. One may think that, if we have high accuracy then our model is best. Yes, accuracy is a great measure but only when you have symmetric datasets where values of false positive and false negatives are almost same. Therefore, you have to look at other parameters to evaluate the performance of your model. For our model, approx. 81% accurate.\n",
    "\n",
    "Accuracy = TP+TN/TP+FP+FN+TN\n",
    "\n",
    "**Precision** - Precision is the ratio of correctly predicted positive observations to the total predicted positive observations. High precision relates to the low false positive rate. \n",
    "\n",
    "Precision = TP/TP+FP\n",
    "\n",
    "**Recall (Sensitivity)** - Recall is the ratio of correctly predicted positive observations to the all observations in actual class - yes. \n",
    "\n",
    "Recall = TP/TP+FN\n",
    "\n",
    "**F1 score** - F1 Score is the weighted average of Precision and Recall. Therefore, this score takes both false positives and false negatives into account. Intuitively it is not as easy to understand as accuracy, but F1 is usually more useful than accuracy, especially if you have an uneven class distribution. Accuracy works best if false positives and false negatives have similar cost. If the cost of false positives and false negatives are very different, it’s better to look at both Precision and Recall. \n",
    "\n",
    "F1 Score = 2*(Recall * Precision) / (Recall + Precision)\n",
    "\n",
    "**Support** - Support is the number of actual occurrences of the class in the specified dataset.\n",
    "\n",
    "\n",
    "**Micro-averaged Precision** is calculated as precision of Total values:\n",
    "\n",
    "$$\n",
    "\\text{Micro-averaged Precision} = \\frac{TP_{total}}{TP_{total} + FP_{total}} \n",
    "$$\n",
    "\n",
    "**Macro-averaged Precision** is calculated as an average of Precisions of all classes\n",
    "\n",
    "**Weighted-averaged Precision** is also calculated based on Precision per class but takes into account the number of samples of each class in the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Example 2 - Naive Bayes Classification on Text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['alt.atheism',\n",
       " 'comp.graphics',\n",
       " 'comp.os.ms-windows.misc',\n",
       " 'comp.sys.ibm.pc.hardware',\n",
       " 'comp.sys.mac.hardware',\n",
       " 'comp.windows.x',\n",
       " 'misc.forsale',\n",
       " 'rec.autos',\n",
       " 'rec.motorcycles',\n",
       " 'rec.sport.baseball',\n",
       " 'rec.sport.hockey',\n",
       " 'sci.crypt',\n",
       " 'sci.electronics',\n",
       " 'sci.med',\n",
       " 'sci.space',\n",
       " 'soc.religion.christian',\n",
       " 'talk.politics.guns',\n",
       " 'talk.politics.mideast',\n",
       " 'talk.politics.misc',\n",
       " 'talk.religion.misc']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_20newsgroups\n",
    "\n",
    "data = fetch_20newsgroups()\n",
    "data.target_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = fetch_20newsgroups(subset='train')\n",
    "test = fetch_20newsgroups(subset='test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "From: rodc@fc.hp.com (Rod Cerkoney)\n",
      "Subject: *$G4qxF,fekVH6\n",
      "Nntp-Posting-Host: hpfcmrc.fc.hp.com\n",
      "Organization: Hewlett Packard, Fort Collins, CO\n",
      "X-Newsreader: TIN [version 1.1 PL8.5]\n",
      "Lines: 15\n",
      "\n",
      "\n",
      "\n",
      "--\n",
      "\n",
      "\n",
      "Regards,\n",
      "Rod Cerkoney\n",
      "                                                        /\\\n",
      "______________________________________________         /~~\\\n",
      "                                                      /    \\\n",
      "  Rod Cerkoney MS 37     email:                      /      \\ \n",
      "  Hewlett Packard         rodc@fc.hp.com        /\\  /        \\  \n",
      "  3404 East Harmony Rd.  Hpdesk:               /  \\/          \\    /\\\n",
      "  Fort Collins, CO 80525  HP4000/UX           /    \\           \\  /  \\\n",
      "_____________________________________________/      \\           \\/    \\__\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(train.data[12])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "model = make_pipeline(TfidfVectorizer(), MultinomialNB())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.fit(train.data, train.target)\n",
    "labels = model.predict(test.data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "fig, ax = plt.subplots(figsize=(10,10))\n",
    "mat = confusion_matrix(test.target, labels)\n",
    "sns.heatmap(mat.T, square=True, annot=True, fmt='d', cbar=False,\n",
    "            xticklabels=train.target_names, yticklabels=train.target_names)\n",
    "plt.xlabel('true label')\n",
    "plt.ylabel('predicted label');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_category(s, train=train, model=model):\n",
    "    pred = model.predict([s])\n",
    "    return train.target_names[pred[0]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'sci.space'"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_category('sending a payload to the ISS')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'alt.atheism'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_category('discussing islam vs atheism')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'comp.graphics'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_category('determining the screen resolution')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Example 3 - Naive Bayes on Breast Cancer Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_breast_cancer\n",
    "cancer = load_breast_cancer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['data', 'target', 'target_names', 'DESCR', 'feature_names', 'filename'])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cancer.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(569, 30)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cancer['data'].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = cancer['data']\n",
    "y = cancer['target']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "StandardScaler(copy=True, with_mean=True, with_std=True)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Now apply the transformations to the data:\n",
    "X_train = scaler.transform(X_train)\n",
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Gaussian Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9230769230769231"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "\n",
    "### Gaussian Naive Bayes\n",
    "clf = GaussianNB()\n",
    "\n",
    "clf.fit(X_train, y_train)\n",
    "clf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Bernoulli Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.965034965034965"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import BernoulliNB\n",
    "\n",
    "### Bernoulli NB\n",
    "clf_mn = BernoulliNB()\n",
    "\n",
    "clf_mn.fit(X_train, y_train)\n",
    "clf_mn.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusions\n",
    "\n",
    "### Advatages:\n",
    "\n",
    "* It is easy and fast to predict class of test data set. It also perform well in multi class prediction\n",
    "\n",
    "\n",
    "* When assumption of independence holds, a Naive Bayes classifier performs better compare to other models like logistic regression and you need less training data.\n",
    "\n",
    "\n",
    "* It perform well in case of categorical input variables compared to numerical variable(s). For numerical variable, normal distribution is assumed.\n",
    "\n",
    "\n",
    "### Disadvantages:\n",
    "\n",
    "* If categorical variable has a category (in test data set), which was not observed in training data set, then model will assign a 0 (zero) probability and will be unable to make a prediction. This is often known as “Zero Frequency”. To solve this, we can use the smoothing technique. One of the simplest smoothing techniques is called Laplace estimation.\n",
    "\n",
    "\n",
    "* Another limitation of Naive Bayes is the assumption of independent predictors. In real life, it is almost impossible that we get a set of predictors which are completely independent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}