Yelp-Rating-Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Yelp Review Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json \n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook will be using the public Yelp dataset:\n",
    "https://www.yelp.com/dataset\n",
    "\n",
    "Here we will be building a support vector machine (SVM) classifier that is modeled using sentiment analysis using texts from reviews. This model will then be able to predict how many star ratings a given review might get."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# may take several minutes to parse 3GB of JSON; alternatively you can skip this cell and use the shortened dataset\n",
    "file = 'dataset/review.json'\n",
    "chunks = pd.read_json(file, lines=True, chunksize=10000)\n",
    "df = pd.concat(chunks, ignore_index=True)\n",
    "df[:300000].to_csv(\"mini_yelp.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the entire dataset contains over 4 million reviews, to save memory, I'm only using the first 300,000 of these reviews, which will be provided along with this notebook. However, using the entire dataset would lead to a more powerful model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "yelp = pd.read_csv(\"mini_yelp.csv\", low_memory=False)\n",
    "yelp = yelp.drop('Unnamed: 0', 1)"
   ]
  },
  {
   "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>business_id</th>\n",
       "      <th>cool</th>\n",
       "      <th>date</th>\n",
       "      <th>funny</th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>useful</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>uYHaNptLzDLoV_JZ_MuzUA</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-07-12</td>\n",
       "      <td>0</td>\n",
       "      <td>VfBHSwC5Vz_pbFluy07i9Q</td>\n",
       "      <td>5</td>\n",
       "      <td>My girlfriend and I stayed here for 3 nights a...</td>\n",
       "      <td>0</td>\n",
       "      <td>cjpdDjZyprfyDG3RlkVG3w</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>uYHaNptLzDLoV_JZ_MuzUA</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-10-02</td>\n",
       "      <td>0</td>\n",
       "      <td>3zRpneRKDsOPq92tq7ybAA</td>\n",
       "      <td>3</td>\n",
       "      <td>If you need an inexpensive place to stay for a...</td>\n",
       "      <td>0</td>\n",
       "      <td>bjTcT8Ty4cJZhEOEo01FGA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>uYHaNptLzDLoV_JZ_MuzUA</td>\n",
       "      <td>0</td>\n",
       "      <td>2015-09-17</td>\n",
       "      <td>0</td>\n",
       "      <td>ne5WhI1jUFOcRn-b-gAzHA</td>\n",
       "      <td>3</td>\n",
       "      <td>Mittlerweile gibt es in Edinburgh zwei Ableger...</td>\n",
       "      <td>0</td>\n",
       "      <td>AXgRULmWcME7J6Ix3I--ww</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>uYHaNptLzDLoV_JZ_MuzUA</td>\n",
       "      <td>0</td>\n",
       "      <td>2016-08-21</td>\n",
       "      <td>0</td>\n",
       "      <td>llmdwOgDReucVoWEry61Lw</td>\n",
       "      <td>4</td>\n",
       "      <td>Location is everything and this hotel has it! ...</td>\n",
       "      <td>0</td>\n",
       "      <td>oU2SSOmsp_A8JYI7Z2JJ5w</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>uYHaNptLzDLoV_JZ_MuzUA</td>\n",
       "      <td>0</td>\n",
       "      <td>2013-11-20</td>\n",
       "      <td>0</td>\n",
       "      <td>DuffS87NaSMDmIfluvT83g</td>\n",
       "      <td>5</td>\n",
       "      <td>gute lage im stadtzentrum. shoppingmeile und s...</td>\n",
       "      <td>0</td>\n",
       "      <td>0xtbPEna2Kei11vsU-U2Mw</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              business_id  cool        date  funny               review_id  \\\n",
       "0  uYHaNptLzDLoV_JZ_MuzUA     0  2016-07-12      0  VfBHSwC5Vz_pbFluy07i9Q   \n",
       "1  uYHaNptLzDLoV_JZ_MuzUA     0  2016-10-02      0  3zRpneRKDsOPq92tq7ybAA   \n",
       "2  uYHaNptLzDLoV_JZ_MuzUA     0  2015-09-17      0  ne5WhI1jUFOcRn-b-gAzHA   \n",
       "3  uYHaNptLzDLoV_JZ_MuzUA     0  2016-08-21      0  llmdwOgDReucVoWEry61Lw   \n",
       "4  uYHaNptLzDLoV_JZ_MuzUA     0  2013-11-20      0  DuffS87NaSMDmIfluvT83g   \n",
       "\n",
       "   stars                                               text  useful  \\\n",
       "0      5  My girlfriend and I stayed here for 3 nights a...       0   \n",
       "1      3  If you need an inexpensive place to stay for a...       0   \n",
       "2      3  Mittlerweile gibt es in Edinburgh zwei Ableger...       0   \n",
       "3      4  Location is everything and this hotel has it! ...       0   \n",
       "4      5  gute lage im stadtzentrum. shoppingmeile und s...       0   \n",
       "\n",
       "                  user_id  \n",
       "0  cjpdDjZyprfyDG3RlkVG3w  \n",
       "1  bjTcT8Ty4cJZhEOEo01FGA  \n",
       "2  AXgRULmWcME7J6Ix3I--ww  \n",
       "3  oU2SSOmsp_A8JYI7Z2JJ5w  \n",
       "4  0xtbPEna2Kei11vsU-U2Mw  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yelp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see above, each row in the reviews dataset consists of a business ID, review ID, user ID, along with how many users have rated each review as 'cool', 'funny', or 'useful', along with the star rating the user gave, and finally, the text of the review. Although we can incorporate other variables by connecting these IDs from the business and user datasets to make a more advanced model, we'll just start by using two variables for our first classification task. We will be using the review texts for our x-value, and be using these to predict star ratings, our y-value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "texts = yelp[\"text\"]\n",
    "stars = yelp[\"stars\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When we train our classification model, we want to make sure that we have an even distribution of possible outcomes, otherwise this may lead to bias. In this case, our outcomes are the five different types of star ratings, so let's check how these are distributed in our dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f900fab3278>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f900fab34a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "matplotlib.style.use('ggplot')\n",
    "% matplotlib inline\n",
    "\n",
    "plt.figure(figsize=(14,3))\n",
    "sns.distplot(stars, kde=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Counter({5: 206714, 4: 119157, 1: 70371, 3: 60514, 2: 43244})\n"
     ]
    }
   ],
   "source": [
    "from collections import Counter\n",
    "print(Counter(stars))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can immediately see that 5-star reviews are hugely overrepresented, while 2 and 3-star reviews tend to be underrepresented. We want an even number of data points across each category, so let's run a quick function that will throw out the reviews in our overrepresented categories until we have a balanced distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    " \n",
    "def balance_classes(xs, ys):\n",
    "    freqs = Counter(ys)\n",
    " \n",
    "    max_allowable = freqs.most_common()[-1][1]\n",
    "    num_added = {clss: 0 for clss in freqs.keys()}\n",
    "    new_ys = []\n",
    "    new_xs = []\n",
    "    for i, y in enumerate(ys):\n",
    "        if num_added[y] < max_allowable:\n",
    "            new_ys.append(y)\n",
    "            new_xs.append(xs[i])\n",
    "            num_added[y] += 1\n",
    "    return new_xs, new_ys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Counter({1: 43244, 2: 43244, 3: 43244, 4: 43244, 5: 43244})\n"
     ]
    }
   ],
   "source": [
    "balanced_x, balanced_y = balance_classes(texts,stars)\n",
    "print(Counter(balanced_y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our dataset is now clean and ready to be prepared for analysis. Machine learning models work using numbers instead of strings, so we need to apply some sort of natural language processing algorithm to turn our collection of texts into meaningful numbers. \n",
    "\n",
    "The most common method is to use the Bag-of-Words (BoW) algorithm, which simply parses the entire dataset and creates a word-count table in the form of a sparse matrix counting up the number of times each word popped up in our collection of texts. This works, but instead we you use the more sophisticated Term Frequency Inverse Document Frequency (TF-IDF) algorithm, which also counts up the number of times a word pops up, but instead of storing just the raw word count in each matrix, it assigns a normalized score to each word based on word frequency. Here is a more detailed resource about the subject: https://deeplearning4j.org/bagofwords-tf-idf\n",
    "\n",
    "The Sklearn library includes a singular function that will account for the tokenization (converting the strings of texts into strings of each individual word) and vectorization (converting these strings into a sparse matrix of TF-IDF scores). All we have to do is specify an 'n-gram range', which allows us to represent multiple words that represent one idea into one element. For example, when we set the n-gram range to 2, instead of counting \"not\" and \"good\" as two words, we'll count \"not good\" as one word."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "vectorizer = TfidfVectorizer(ngram_range=(1,2)) # may take a few minutes\n",
    "vectors = vectorizer.fit_transform(balanced_x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Like we do with all machine learning models, we now need to partition our data into the training set (the section of the data that the model learns from) and the test set (the section of the data that is NOT used to train the model and instead is used to test the model's accuracy). We will use a test size of 0.33, meaning we'll train the model on 2/3 of the dataset, and test it on the last 1/3 of the remaining data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "x_train, x_test, y_train, y_test = train_test_split(vectors, balanced_y, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we can start training our machine learning model. We will be using a support vector machine (SVM) classifier: https://en.wikipedia.org/wiki/Support_vector_machine\n",
    "\n",
    "We don't need to necessarily know the math behind it to apply this model (although it certainly is useful to know). All we need to do is pass in a set of features from our training set (x_train, our collection of vectorized review) and a set of labels from our training set (y_train, the star ratings associated with each review) into the LinearSVC class provided to us by the Scikit-learn package."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearSVC(C=1.0, class_weight=None, dual=True, fit_intercept=True,\n",
       "     intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
       "     multi_class='ovr', penalty='l2', random_state=None, tol=0.0001,\n",
       "     verbose=0)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import LinearSVC\n",
    "classifier = LinearSVC()\n",
    "classifier.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our model is now complete. \n",
    "\n",
    "We can try applying the model to some of the examples in our test set to see how well it classifies them. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2, 3, 4, 1, 5, 1, 5, 5, 1, 4, 2, 3, 1, 3, 3, 2, 1, 2, 2, 3]\n",
      "[1, 3, 2, 1, 4, 3, 5, 5, 3, 4, 3, 3, 1, 4, 3, 2, 1, 3, 1, 4]\n"
     ]
    }
   ],
   "source": [
    "predictions = classifier.predict(x_test)\n",
    "print(list(predictions[:20]))\n",
    "print(y_test[:20])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first list represents the model's predictions of star ratings for the first twenty reviews, while the second list represents the actual star ratings of the first twenty reviews. We can also calculate the accuracy score when the model is applied to the entire test set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.611424887531\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "print(accuracy_score(y_test, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our model guesses the correct star rating with a 61% accuracy. This is actually pretty decent when considering that  the model is choosing between five different categories. Imagine if you were given a lukewarm review and had to choose whether or not that review gave a 2-star, 3-star, or 4-star rating. That would be difficult task even for a human. We can construct a confusion matrix which will tell us how often one category of rating may be confused with another."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[11188  2343   411    99   129]\n",
      " [ 3509  7126  3028   525   215]\n",
      " [  798  2958  7092  2962   588]\n",
      " [  209   535  2606  7387  3471]\n",
      " [  160   117   393  2670 10834]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "print(confusion_matrix(y_test, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The x-axis represents the SVM classifier predicted, while the y-axis represents what the actual star rating was. For example, the first row represents what predicted as having one-star while the columns represent how many stars the review actually received, and the second row represents what the SVM predicted as having two-stars, third-row as three-stars, and so on. \n",
    "\n",
    "So we can see using this that the model is pretty accurate for guessing extremes (1 or 5-stars), but struggles more with determining whether or not a more lukewarm review should be classified as either a 2-star, 3-star, or 4-star review.\n",
    "\n",
    "To gain more insight into how our SVM model makes its predictions, we can look into how our model calculates its prediction by seeing what features (in this case, the features are words) are weighted highest and lowest in determining how to classify each review. We can order all the words by feature weight."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Highest Weighted Words for N-Star Reviews:\n",
      "1: avoid, joke, worse, not recommend, could give, rude, stay away, never, no stars, scam, zero, never again, awful, disgusting, zero stars, will never, terrible, horrible, one star, worst\n",
      "2: subpar, so so, unfortunately, won be, unimpressed, bland, not, to like, at best, mediocre, stars because, not very, disappointed, not worth, experienced better, second star, disappointing, meh, two star, two stars\n",
      "3: not great, to love, alright, not so, okay, aber, back but, other hand, though, three star, average, an ok, but, ok, decent, stars, again but, not bad, however, three stars\n",
      "4: even better, chandelier, the better, goodness, not fancy, ru, tasty, off star, above average, shakes, makes up, nice, delicious, star because, ikea, really good, my only, star off, good, four stars\n",
      "5: gem, honest, wonderful, hands down, every, love, be disappointed, thanks, so good, not disappointed, incredible, perfect, great, outstanding, excellent, not only, fantastic, awesome, amazing, best\n"
     ]
    }
   ],
   "source": [
    "def print_top20_best(vectorizer, clf, class_labels):\n",
    "    feature_names = vectorizer.get_feature_names()\n",
    "    for i, class_label in enumerate(class_labels):\n",
    "        top20 = np.argsort(clf.coef_[i])[-20:]\n",
    "        print(\"%s: %s\" % (class_label,\n",
    "              \", \".join(feature_names[j] for j in top20)))\n",
    "        \n",
    "print(\"Highest Weighted Words for N-Star Reviews:\")\n",
    "print_top20_best(vectorizer, classifier, classifier.classes_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Lowest Weighted Words for N-Star Reviews:\n",
      "1: stars, great, two stars, stars because, delicious, excellent, amazing, good, awesome, best, love, not bad, but, decent, friendly, fantastic, nice, tasty, loved, enjoyed\n",
      "2: three stars, delicious, will never, great, amazing, perfect, worst, excellent, zero stars, awesome, not too, perfectly, if could, could give, fantastic, four stars, never again, solid, perfection, be disappointed\n",
      "3: two stars, one star, awesome, not only, gem, worth it, ok but, terrible, can wait, even, below average, horrible, delicious, be back, worst, highly, love this, experienced better, perfect, recommend\n",
      "4: mediocre, three stars, bland, two stars, disappointing, worst, not worth, ok, nothing, terrible, meh, overpriced, disappointed, not, okay, poor, horrible, average, nice but, that about\n",
      "5: worst, however, not, horrible, rude, decent, terrible, disappointing, only reason, slow, ok, but, awful, okay, average, not worth, pretty, at best, wasn, overpriced\n"
     ]
    }
   ],
   "source": [
    "def print_top20_worst(vectorizer, clf, class_labels):\n",
    "    feature_names = vectorizer.get_feature_names()\n",
    "    for i, class_label in enumerate(class_labels):\n",
    "        top20 = np.argsort(clf.coef_[i])[:20]\n",
    "        print(\"%s: %s\" % (class_label,\n",
    "              \", \".join(feature_names[j] for j in top20)))\n",
    "\n",
    "print(\"Lowest Weighted Words for N-Star Reviews:\")\n",
    "print_top20_worst(vectorizer, classifier, classifier.classes_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we can also generate these in the form of a Matplotlib/Seaborn plot that graphs the highest and lowest ranking words by their coefficients. We'll generate a plot for each classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8fcac905c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8ff8e29320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8feaf92ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8fe45e9cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8feb105518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_coefficients(classifier, feature_names, stars, top_features=15):\n",
    "    coef = classifier.coef_[stars - 1]\n",
    "    top_positive_coefficients = np.argsort(coef)[-top_features:]\n",
    "    top_negative_coefficients = np.argsort(coef)[:top_features]\n",
    "    top_coefficients = np.hstack([top_negative_coefficients, top_positive_coefficients])\n",
    "\n",
    "    plt.figure(figsize=(15, 5))\n",
    "    plt.bar(np.arange(2 * top_features), coef[top_coefficients])\n",
    "    labels = []\n",
    "    for i in top_coefficients:\n",
    "        labels.append(feature_names[i])\n",
    "    plt.xticks(np.arange(0, 1 + 2 * top_features), labels, rotation=70)\n",
    "    plt.title(\"Least and Most Likely Words/Phrases for \" + str(stars) + \"-Star Reviews\")\n",
    "    plt.show()\n",
    "\n",
    "for i in range(1,6):\n",
    "    plot_coefficients( classifier, vectorizer.get_feature_names(), i )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also chart all of the different coefficients at once to get a holistic picture of how our model uses different words to predict review type."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_all_coefficients(classifier, feature_names):\n",
    "    # gets indices of top 15 words from 5-Star Reviews, 3-Star Reviews, and 1-Star Reviews respectively\n",
    "    coef = classifier.coef_[4]\n",
    "    top_coefficients = list(np.argsort(coef)[-15:])[::-1]\n",
    "    coef = classifier.coef_[2]\n",
    "    top_coefficients = top_coefficients + list(np.argsort(coef)[-15:])\n",
    "    coef = classifier.coef_[0]\n",
    "    top_coefficients = top_coefficients + list(np.argsort(coef)[-15:])\n",
    "    \n",
    "    # creates arrays representing coefficients obtained from indices of top words\n",
    "    one_star_coefficients = classifier.coef_[0][top_coefficients]\n",
    "    two_star_coefficients = classifier.coef_[1][top_coefficients]\n",
    "    three_star_coefficients = classifier.coef_[2][top_coefficients]\n",
    "    four_star_coefficients = classifier.coef_[3][top_coefficients]\n",
    "    five_star_coefficients = classifier.coef_[4][top_coefficients]\n",
    "    \n",
    "    # creates array of words using indices from top_coefficients\n",
    "    feature_names = vectorizer.get_feature_names()\n",
    "    words = []\n",
    "    for i in top_coefficients:\n",
    "        words.append(feature_names[i])\n",
    "      \n",
    "    # creates pandas dataframe of coefficients and associated words\n",
    "    scores_list = pd.DataFrame({\"words\": words, \n",
    "                                \"One Star Reviews\":   one_star_coefficients,\n",
    "                                \"Two Star Reviews\":   two_star_coefficients,\n",
    "                                \"Three Star Reviews\": three_star_coefficients,\n",
    "                                \"Four Star Reviews\":  four_star_coefficients,\n",
    "                                \"Five Star Reviews\":  five_star_coefficients})\n",
    "    scores_list = scores_list[[\"words\",\n",
    "                               \"One Star Reviews\",\n",
    "                               \"Two Star Reviews\",\n",
    "                               \"Three Star Reviews\",\n",
    "                               \"Four Star Reviews\",\n",
    "                               \"Five Star Reviews\"]]\n",
    "    scores_list = pd.melt(scores_list, id_vars=[\"words\"], \n",
    "                        var_name=\"Review Type\", value_name=\"TF-IDF Coefficient\")\n",
    "    \n",
    "    # creates chart in seaborn\n",
    "    color_palettee = sns.color_palette(\"PuBuGn_d\", 5)\n",
    "    sns.set_palette(color_palettee)\n",
    "    plt.figure(figsize=(10,30))\n",
    "    sns.barplot(orient=\"h\", y=\"words\", x=\"TF-IDF Coefficient\", hue=\"Review Type\", data=scores_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8ff5373b00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_all_coefficients(classifier, vectorizer.get_feature_names())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above, we can see what qualities are most commonly used in high- or low-ranking establishment. High-ranking establishments are usually described as \"best\", \"amazing\", or \"awesome\". Low-ranking establishments are usually associated with the words \"horrible\", \"worst\", \"disgusting\", \"scam\", and \"rude\". The words \"decent\" and \"average\" are mostly used when describing average-rated establishments."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An accuracy score of 0.61 is decent, but can we do better than that? What are some ways we can improve the accuracy of our model further? We can:\n",
    "- Feed the model more data. Since we only used 300,000 reviews out of the dataset, using all 4 million reviews from the original dataset would likely lead to better results.\n",
    "- Add more features to the model. We made our classifier based purely on TF-IDF scores from the text. We could try adding in more variables such as the number of 'cool', 'useful', or 'funny' scores that a review gets to see if that changes things. We could also try getting different features from the text, such as calculating review length and seeing if that influences the data.\n",
    "- Break model down by category. If we combine the 'review' dataset with the 'business' data, we get access to what category each review is in, what type of business, etc. We may have better results by adding in information about the business to our model.\n",
    "- Redefine the problem. Trying to identify whether a lukewarm review is 2-, 3-, or 4-star can be an impossible task even for a human. Perhaps if we narrowed down our model to only identifying whether or not a review was 'positive', 'negative', or 'neutral' would be more feasible and yield better results."
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}